Skip to content

T
Toy-MT-Introduction
Commit ee3019bf by 曹润柘
Options

update chapter1-3 to the latest

parent 5742ff2f
正在显示 包含 2219 行增加 和 1466 行删除
Book/Chapter1/Figures/figure-Required-parts-of-MT.tex
......@@ -21,7 +21,7 @@
{
\begin{scope}[yshift=-1.0in,xshift=2.8in]
\node [] (part2label) at (0,0) {\includegraphics[scale=0.33]{./Chapter1/Figures/figure-truck.jpg}};
\node [anchor=north west] (part2text) at ([xshift=-1em,yshift=0em]part2label.south west) {\begin{tabular}{l}{\color{red}{系统}}：利用资源完成\\完成自动翻译的程序\end{tabular}};
\node [anchor=north west] (part2text) at ([xshift=-1em,yshift=0em]part2label.south west) {\begin{tabular}{l}{\color{red}{系统}}：利用资源完成\\自动翻译的程序\end{tabular}};
\end{scope}
}
......
Book/Chapter1/chapter1.tex
......@@ -333,12 +333,10 @@
\parinterval BLEU的计算首先考虑待评价译文中$n$-gram在参考答案中的匹配率，称为准确率。其计算方法如下：
\begin{equation}
\begin{split}
\begin{eqnarray}
\textrm{P}_n=\frac{\textrm{Count}_\textrm{hit}}{\textrm{Count}_{\textrm{output}}}
\label{eq:matching-rate}
\end{split}
\end{equation}
\end{eqnarray}
\parinterval 其中$\textrm{Count}_{\textrm{hit}}$表示系统输出的译文中$n$-gram在参考答案中命中的次数，$\textrm{Count}_{\textrm{output}}$\\表示系统输出的译文中总共有多少$n$-gram。为了避免同一个词被重复计算，BLEU的定义中使用了截断的方式定义$\textrm{Count}_{\textrm{hit}}$和$\textrm{Count}_{\textrm{output}}$。例如：
......@@ -352,34 +350,28 @@ Candidate：the the the the
\parinterval译文整体的准确率等于各$n$-gram的加权平均：
\begin{equation}
\begin{split}
\begin{eqnarray}
P_{\textrm{avg}}=\exp(\sum_{n=1}^Nw_n\cdot \log{P_n})
\label{eq:weighted-average}
\end{split}
\end{equation}
\end{eqnarray}
\parinterval但是，该方法更倾向于对短句子打出更高的分数，一个极端的例子译文只有很少的几个词，但是都命中答案，准确率很好可显然不是好的译文。因此，BLEU引入惩罚因子（Brevity Penalty, BP）的概念，对短句进行惩罚，
\begin{equation}
\begin{split}
\begin{eqnarray}
\textrm{BP}=
\begin{cases}
1& c>r\\
e^{(1-\frac{r}{c})}& c<r
\end{cases}
\label{eq:brevity-penalty}
\end{split}
\end{equation}
\end{eqnarray}
\noindent 其中$c$ : 译文的句子长度，$r$ : 参考译文的句子长度，所以最终BLEU的计算公式为：
\begin{equation}
\begin{split}
\begin{eqnarray}
\textrm{BLEU}=\textrm{BP} \cdot \exp(\sum_{i=1}^{N}w_n \cdot \log{\textrm{P}_n})
\label{eq:BLUE}
\end{split}
\end{equation}
\end{eqnarray}
\parinterval从机器翻译的发展来看，BLEU的意义在于它给系统研发人员提供了一种简单、高效、可重复的自动评价手段，在研发机器翻译系统时可以不需要依赖人工评价。同时，BLEU也有很多创新之处，包括引入$n$-gram的匹配，截断计数和短句惩罚等等，包括NIST等很多评价指标都是受到BLEU的启发。当然， BLEU也并不完美，甚至经常被人诟病，比如，它需要依赖参考译文，而且评价结果有时与人工评价不一致，同时BLEU评价只是单纯的从匹配度的角度思考翻译质量的好坏，并没有真正考虑句子的语义是否翻译正确。但是，毫无疑问，BLEU仍然是机器翻译中最常用的评价方法。在没有找到更好的替代方案之前，BLEU还是机器翻译研究所使用的标准评价指标。
......@@ -387,12 +379,10 @@ e^{(1-\frac{r}{c})}& c<r
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\parinterval TER是Translation Edit Rate的缩写，是一种基于距离的评价方法，用来评定机器翻译结果的译后编辑的工作量\cite{snover2006study}。这里，距离被定义为两个序列之间其中一个转换成另一个序列所需要的最少编辑操作次数。操作次数越多，距离越大，序列之间的相似性越低，相反距离越小，表示一个句子越容易改写成另一个句子，序列之间的相似性越高。TER使用的编辑操作包括：增加，删除，替换和移位，其中增加，删除，替换操作计算得到的距离被称为编辑距离，并根据错误率的形式给出评分：
\begin{equation}
\begin{split}
\begin{eqnarray}
\textrm{score}=\frac{\textrm{edit}(c,r)}{l}
\label{eq:score-based-on-error-rate}
\end{split}
\end{equation}
\end{eqnarray}
\noindent 其中$\textrm{edit}(c,r)$是指机器翻译生成的候选译文$c$和参考译文$r$之间的距离，$l$是归一化因子，通常为参考译文的长度。在距离计算中所有的操作的代价都为1。在计算距离时，优先考虑移位操作，再计算编辑距离，也就是增加、删除和替换操作的次数。直到增加、移位操作无法减少编辑距离时，将编辑距离和移位操作的次数累加得到TER计算的距离。例如:
......@@ -539,31 +529,31 @@ His house is on the south bank of the river.
\vspace{0.5em}
\item Zoph：Zoph是由Information Sciences Institute 的Barret Zoph 等人使用C++语言开发的系统。Zoph在多个GPU 上既可以训练序列模型（例如语言模型），也可以训练序列到序列的模型（例如神经机器翻译模型），且可通过参数调整网络的层数。当训练NMT系统时，Zoph也支持了多源输入，即在输入源语句子时可同时输入其一种译文。该系统由于使用C++语言，所以有运行速度快的特点。\url{https://github.com/isi-nlp/Zoph\_RNN}
\vspace{0.5em}
\item Fairseq：Fairseq是由facebook的AI研究小组开发的，基于Torch框架的用以解决序列到序列问题的工具包，其中包括基于卷积神经网络的机器翻译模型，基于循环神经网络的机器翻译模型，以及Transformer模型等。https://github.com/facebookresearch/fairseq
\item Fairseq：Fairseq是由facebook的AI研究小组开发的，基于Torch框架的用以解决序列到序列问题的工具包，其中包括基于卷积神经网络的机器翻译模型，基于循环神经网络的机器翻译模型，以及Transformer模型等。\\ https://github.com/facebookresearch/fairseq
\vspace{0.5em}
\item Tensor2Tensor：Tensor2Tensor是由谷歌推出的，基于TensorFlow框架的Transformer系统。该系统与之前的使用循环神经网络或卷积神经网络结构不同，而是使用self-attention机制以及最简单的前馈神经网络构成的。得益于Transformer的网络结构，可使得系统可以在多个GPU上并行运行，大大加快了训练的速度。该系统由于没有循环等复杂的运算，故训练和解码都比循环神经网络快。此外，目前该系统的翻译效果比循环神经网络结构的神经机器翻译系统更好一点。\url{https://github.com/tensorflow/tensor2tensor}
\item Tensor2Tensor：Tensor2Tensor是由谷歌推出的，基于TensorFlow框架的Transfor\\ -mer系统。该系统与之前的使用循环神经网络或卷积神经网络结构不同，而是使用self-attention机制以及最简单的前馈神经网络构成的。得益于Transformer的网络结构，可使得系统可以在多个GPU上并行运行，大大加快了训练的速度。该系统由于没有循环等复杂的运算，故训练和解码都比循环神经网络快。此外，目前该系统的翻译效果比循环神经网络结构的神经机器翻译系统更好一点。\url{https://github.com/tensorflow/tensor2tensor}
\vspace{0.5em}
\item OpenNMT：OpenNMT系统是由Harvard NLP（哈佛大学自然语言处理研究组）开源的，基于Torch框架的神经机器翻译系统。OpenNMT系统的早期版本使用Lua 语言编写，现在也扩展到了TensorFlow和PyTorch，设计简单易用，易于扩展，同时保持效率和翻译精度。其接口简单通用，只需源/目标文件。可以在GPU 上快速高性能训练，且优化了内存。此外，OpenNMT可扩展到其他序列生成任务，例如文本摘要和多模态翻译等。\url{https://github.com/OpenNMT/OpenNMT}
\vspace{0.5em}
\item 斯坦福NMT开源代码库：斯坦福大学自然语言处理组（Stanford NLP）发布了一片文章，总结了该研究组在神经机器翻译上的研究信息，同时他们实现了当前最佳结果的代码库。斯坦福NMT开源代码库包括了3种NMT系统，分别为基于字词混合的hybrid NMT，基于注意力机制的attention-based NMT以及通过剪枝方式压缩模型的pruning NMT。\url{https://nlp.stanford.edu/projects/nmt/}
\vspace{0.5em}
\item THUMT：清华大学NLP团队实现的Transformer等模型、支持多GPU训练和解码、分布式训练。\url{https://github.com/THUNLP-MT/THUMT}
\item THUMT：清华大学NLP团队实现的Transformer等模型、支持多GPU训练和解码、分布式训练。主要基于Tensorflow和Theano实现，其中Theano版本包含了RNNsearch模型，训练方式包括MLE（Maximum Likelihood Estimate）, MRT（minimum risk training）, SST（semi-supervised training）。优化器包括：SGD, Adadelta, Adam，当前Theano版本已不再更新。官方推荐使用Tensorflow版本，其实现了Seq2Seq, RNNsearch, Transformer三种基本模型，支持MLE训练方式以及Adam优化器。\url{https://github.com/THUNLP-MT/THUMT}
\vspace{0.5em}
\item NiuTrans.NMT：东北大学NLP团队基于NiuTensor工具集实现的Transformer 模型以及前馈网络语言模型。\url{https://github.com/NiuTrans/NiuTensor}
\item NiuTrans.NMT：由东北大学NLP团队基于NiuTensor工具集实现，支持前馈网络，循环神经网络，Transformer等结构，支持语言模型，序列标注，机器翻译等任务。支持机器翻译GPU与CPU训练及解码。其小巧易用，为NLP开发人员提供快速二次开发基础。\url{https://github.com/NiuTrans/NiuTensor}
\vspace{0.5em}
\item MARIANNMT：主要由微软翻译团队搭建，使用纯C++实现的用于GPU/CPU\\训练和解码的引擎。\url{https://marian-nmt.github.io/}
\item MARIANNMT：主要由微软翻译团队搭建，使用纯C++实现的用于GPU/CPU训练和解码的引擎，支持多GPU训练和批量解码，最小限度依赖第三方库，静态编译一次之后，复制其二进制文件就能在其他平台使用。\url{https://marian-nmt.github.io/}
\vspace{0.5em}
\item Sockeye：Awslabs开发的一种序列到序列的神经机器翻译框架。\url{https://awslabs.github.io/sockeye/}
\item Sockeye：由Awslabs开发得神经机器翻译框架。其中支持RNN Search，Transfor\\ -mer,CNN等翻译模型，同时还有提供了从图片翻译到文字的模块。提供了WMT 德英新闻翻译，领域适应任务，多语言零资源翻译任务的教程。\url{https://awslabs.github.io/sockeye/}
\vspace{0.5em}
\item CytonMT：由NICT开发的一种用C++实现的高效神经机器翻译开源工具包。\url{https://github.com/arthurxlw/cytonMt}
\item CytonMT：由NICT开发的一种用C++实现的高效神经机器翻译开源工具包。主要支持Transformer模型，支持了SGD和Adam优化器，一些常用的训练方法以及解码方法。\url{https://github.com/arthurxlw/cytonMt}
\vspace{0.5em}
\item OpenSeq2Seq：由NVIDIA团队开发的模块化架构，支持混合精度训练，基于Horovod的快速分布式训练。\url{https://nvidia.github.io/OpenSeq2Seq/html/index.html}
\item OpenSeq2Seq：由NVIDIA团队开发的基于tensorflow的模块化架构，用于序列到序列的模型，允许从可用组件中组装新模型,支持混合精度训练，利用NVIDIA Volta / Turing GPU中的Tensor核心，基于Horovod的快速分布式训练，支持多gpu，多节点多模式。\url{https://nvidia.github.io/OpenSeq2Seq/html/index.html}
\vspace{0.5em}
\item Nmtpytorch：由勒芒大学语言实验室构建的端到端的神经结构。\url{https://github.com/lium-lst/nmtpytorch}
\item Nmtpytorch：由勒芒大学语言实验室发布的基于序列到序列框架的神经网络翻译模型，Nmtpytorch的核心部分依赖于Numpy，Pytorch和tqdm。其允许训练各种端到端神经体系结构，包括但不限于神经机器翻译，图像字幕和自动语音识别系统。\url{https://github.com/lium-lst/nmtpytorch}
\vspace{0.5em}
\item DL4MT：由Cho Lab at NYU CS and CDS团队构建的多编码器、多解码器或多路NMT模型。\url{https://github.com/nyu-dl/dl4mt-multi}
\item DL4MT：由Cho Lab at NYU CS and CDS团队构建的多编码器、多解码器或多路NMT模型。改项目主要使用Theano 构建计算图，吸取了其他NMT系统搭建的经验，构建了调度器来管理调度多个数据流从而训练多个计算图，从而加快训练速度。同时该团队还提供了对应的学习材料，进一步讲解整个项目的细节。\url{https://github.com/nyu-dl/dl4mt-multi}
\vspace{0.5em}
\item ModerNMT：由modernmt团队搭建的基于Fairseq变换器模型的上下文感知、增量分布式通用神经机器翻译。\url{https://github.com/ModernMT/MMT}
\item ModerNMT：由modernmt团队搭建的基于Fairseq Tranformer模型的上下文感知、增量和分布式通用的神经机器翻译技术，易于使用和扩展。它通过将所有可用的项目/客户数据和翻译记忆库集中在一个文件夹中进行训练，通过提供要翻译的句子和可选的一些上下文文本来查询。\url{https://github.com/ModernMT/MMT}
\end{itemize}
\vspace{0.5em}
......@@ -590,7 +580,7 @@ His house is on the south bank of the river.
\parinterval以上机器翻译评测各自有不同的特点，NIST最近几年更加关注稀缺资源翻译相关问题；NTCIR 在评估方式上纳入了时间、多语种评估等手段；WAT倾向于评测亚洲相关语言的翻译；CCMT以汉语为核心，并支持国内许多少数民族语言；WMT面向欧洲语系，现在也延伸到汉语，语种范围广，评测类型丰富；IWSLT针对语音对话的翻译相关问题进行评测。
\parinterval以上评测数据大多可以从评测网站上下载，此外部分数据也可以从LDC（Linguistic Data Consortium）上申请，网址为\url{https://www.ldc.upenn.edu/}。ELRA（European Language Resources Association）上也有一些免费的语料库供研究使用，其官网为\url{http://www.elra.info/}。{\red 更多机器翻译的语料信息可参看附录？？}。
\parinterval以上评测数据大多可以从评测网站上下载，此外部分数据也可以从LDC（Lingu \\ -istic Data Consortium）上申请，网址为\url{https://www.ldc.upenn.edu/}。ELRA（Euro \\ -pean Language Resources Association）上也有一些免费的语料库供研究使用，其官网为\url{http://www.elra.info/}。{\red 更多机器翻译的语料信息可参看附录？？}。
\parinterval从机器翻译发展的角度看，这些评测任务给相关研究提供了基准数据集，使得不同的系统都可以在同一个环境下进行比较和分析，进而建立了机器翻译研究所需的实验基础。此外，这些评测任务也使得研究者可以第一时间了解机器翻译研究的最新成果，比如，有多篇ACL最佳论文的灵感就来自当年参加机器翻译评测任务的系统。
......@@ -624,8 +614,26 @@ His house is on the south bank of the river.
\vspace{0.5em}
\item AMTA，全称Association for Machine Translation in the Americas。AMTA会议汇聚了学术界、产业界和政府的MT研究人员、开发人员和用户，让工业界和学术界进行交流。
\vspace{0.5em}
\item {\red CCL}
\item {\red NLPCC}
\item {\red CCMT}
\item CCL，全称Conference on Computational Linguistics，中文为中国计算语言学大会。中国计算语言学大会创办于1991年，由中国中文信息学会计算语言学专业委员会负责组织。经过20余年的发展，中国计算语言学大会已成为国内自然语言处理领域权威性最高、规模和影响最大的学术会议。作为中国中文信息学会（国内一级学会）的旗舰会议，CCL聚焦于中国境内各类语言的智能计算和信息处理，为研讨和传播计算语言学最新学术和技术成果提供了最广泛的高层次交流平台。
\vspace{0.5em}
\item NLPCC，全名The CCF International Conference on Natural Language Processing and Chinese Computing。NLPCC 是由中国计算机学会（CCF）主办的 CCF 中文信息技术专业委员会年度学术会议，同时也是CCF推荐C类会议，专注于自然语言处理及中文计算领域的学术和应用创新。会议自 2012 年开始举办，主要活动有主题演讲、论文报告、技术测评、专家论坛、技术展示。
\vspace{0.5em}
\item CCMT，全名The China Conference on Machine Translation，中国机器翻译研讨会，由中国中文信息学会主办，旨在为国内外机器翻译界同行提供一个平台，加强国内外同行的学术交流，促进中国机器翻译事业。CCMT不仅是国内机器翻译领域最具影响力、最权威的学术和评测活动，而且也代表着汉语与民族语言翻译技术的最高水准，对民族语言技术发展具有重要意义。
\vspace{0.5em}
\end{itemize}
\vspace{0.5em}
Book/Chapter2/Figures/figure-29.tex deleted 100644 → 0
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
\begin{tikzpicture}
\begin{scope}[grow'=up,level distance=23pt]
\Tree[.\node(t1){IP};
[.\node(t2){NP}; \edge[roof]; \node(t3){$\textrm{中国}_{1}\textrm{明星}_{2}\textrm{艺术团}_{3}$}; ]
[.\node(t4){VP};
[.\node(t5){BA};\node(t9){$\textrm{将}_{4}$};]
[.\node(t6){NP};\edge[roof]; \node(t10){$\textrm{一}_{5}\textrm{台}_{6}\textrm{精彩}_{7}\textrm{的}_{8}\textrm{京剧}_{9}\textrm{歌舞}_{10}$};]
[.\node(t7){VP};
[.\node(t11){VV}; \node(t12){$\textrm{呈现}_{11}$};]
]
[.\node(t8){PP};
[.\node(t13){P}; \node(t15){$\textrm{给}_{12}$};]
[.\node(t14){NP};\edge[roof]; \node(t16){$\textrm{香港}_{13}\textrm{观众}_{14}$};]
]
]
[..
]
]
\end{scope}
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-Example-of-word-segmentation-based-on-dictionary.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
......@@ -142,8 +109,6 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-MT-system-as-a-black-box.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
......@@ -119,8 +83,4 @@
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-MT=language-analysis+translation-engine.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
......@@ -108,8 +71,6 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-a-simple-pre-processing-process.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
......@@ -44,7 +9,7 @@
\vspace{0.5em}
\begin{tikzpicture}
{\footnotesize
{\small
\node [ugreen] (input) at (0,0) {猫喜欢吃鱼};
\node [draw,thick,anchor=west,ublue] (preprocessing) at ([xshift=1em]input.east) {分词系统};
\node [ugreen,anchor=west] (mtinput) at ([xshift=1em]preprocessing.east) {猫/喜欢/吃/鱼};
......@@ -58,7 +23,3 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-analysis-of-sentence-participle&syntactic.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
......@@ -45,8 +6,8 @@
\begin{tikzpicture}
\begin{scope}[scale=0.8,xshift=0.9in,yshift=-0.87in,level distance=20pt,sibling distance=-1pt,grow'=up]
{\scriptsize
\begin{scope}[scale=1.0,xshift=0.9in,yshift=-0.87in,level distance=20pt,sibling distance=-1pt,grow'=up]
{
\Tree[.\node(sn0){IP};
[.\node(sn1){NP};
[.\node(sn2){NN}; \node(sw1){猫}; ]
......@@ -62,7 +23,3 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-cross-type-word-segmentation-ambiguity.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
......@@ -95,7 +60,6 @@
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/evaluation-of-probability-for-grammar.tex → Book/Chapter2/Figures/figure-evaluation-of-probability-for-grammar.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
\begin{tikzpicture}
......@@ -130,7 +94,5 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/example-of-derivation.tex → Book/Chapter2/Figures/figure-example-of-derivation.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
\begin{minipage}[t]{0.38\linewidth}
......@@ -204,7 +168,5 @@
\end{minipage}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-example-of-zh-syntactic-analysis.tex 0 → 100644
%%% outline
%-------------------------------------------------------------------------
\begin{tikzpicture}
\node[anchor=west] (input) at (0,0) {\textbf{ \scriptsize{源句语法树}}};
\node[anchor=north west] (rule) at ([xshift=-14.5em,yshift=11.5em]input.south west) {\scriptsize{句法翻译规则:}};
\node[anchor=north west] (rule2) at ([yshift=0.2em]rule.south west) {\scriptsize{(VP BA(将) $x_1$:NP $x_2$:VP PP(P(给) $x_3$:NP))}};
\node[anchor=north west] (rule3) at ([yshift=0.2em]rule2.south west) {\scriptsize{$\to$ $x_2$ $x_1$ to $x_3$ }};
\begin{scope}[scale = 0.7, grow'=up, sibling distance=6pt, level distance=35pt, xshift=3.6em, yshift=1.0em]
\Tree[.\node(tn1){IP};
[.\node(tn2){NP}; \edge[roof]; \node[](seg1){中国$_1$ 明星$_2$ 艺术团$_3$}; ]
[.\node(tn3){VP};
[.\node(tn4){BA}; \node[fill=red!20!white](seg2){将$_4$}; ]
[.\node(tn5){NP}; \edge[roof]; \node[fill=blue!20!white](seg3){一$_5$ 台$_6$ 精彩$_7$ 的$_8$ 京剧$_9$ 歌舞$_{10}$}; ]
[.\node(tn6){VP};
[.\node(tn7){VV}; \node[fill=green!20!white](seg4){呈现$_{11}$}; ]
]
[.\node(tn8){PP};
[.\node(tn9){P}; \node[fill=red!20!white](seg5){给$_{12}$}; ]
[.\node(tn10){NP}; \edge[roof]; \node[fill=purple!20!white](seg6){香港$_{13}$ 观众$_{14}$}; ]
]
]
[.\node(tn11){.}; ]
]
\end{scope}
\draw [-,dotted,very thick,blue] (tn3.south west) -- (tn4.south west) -- (seg2.north west) -- (seg2.north east) -- ([xshift=0.0em]tn4.north east) -- ([xshift=0.8em]tn6.north east) -- (seg5.north west) -- (seg5.north east) -- ([yshift=-1.2em]seg5.south east) -- (tn10.north east) -- (tn10.south east) -- (tn8.south east) -- (tn3.south east) -- (tn3.south west);
\draw[<->,dotted,blue,thick] ([yshift=-0.3em]seg2.south west) .. controls +(west:0.7) and +(south:0.5) .. ([xshift=-1.8em]rule2.south);
\end{tikzpicture}
%---------------------------------------------------------------------
Book/Chapter2/Figures/figure-examples-of-Chinese-word-segmentation-based-on-1-gram-model.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
......@@ -79,7 +43,7 @@
{\footnotesize
{
\node [anchor=west] (label1) at (0,6em) {实际上，通过学习我们得到了一个分词模型P($\cdot$)，给定任意的};
\node [anchor=north west] (label1part2) at ([yshift=0.5em]label1.south west) {分词结果$W=w_1 w_2...w_n$，都能通过P($W$)=$\textrm{P}(w_1) \cdot \textrm{P}(w_2) \cdot ... \cdot \textrm{P}(w_n)$};
\node [anchor=north west] (label1part2) at ([yshift=0.5em]label1.south west) {分词结果$W=w_1 w_2...w_n$，都能通过P($W$)=$\textrm{P}(w_1) \cdot \textrm{P}(w_2) \cdot ... \cdot \textrm{P}(w_n)$ \; \, };
\node [anchor=north west] (label1part3) at ([yshift=0.5em]label1part2.south west) {计算这种分词的概率值};
}
......@@ -165,8 +129,5 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-full-probability-word-segmentation-1.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
......@@ -83,8 +48,5 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-full-probability-word-segmentation-2.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
......@@ -77,7 +43,6 @@
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-full-probability-word-segmentation-3.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
\begin{tikzpicture}
\begin{scope}[yshift=-2.8cm,xshift=-0.5em]
\node [anchor=west] (label1) at (0,0) {总词数：$6 + 8 + 5 = 20$};
\node [anchor=north west] (p1) at (label1.south west) {$\textrm{P}('\textrm{很}')=1/20=0.05$};
\node [anchor=north west] (p2) at (p1.south west) {$\textrm{P}('\textrm{。}')=3/20=0.15$};
\node [anchor=north west] (p3) at (p2.south west) {$\textrm{P}('\textrm{确实}')=1/20=0.05$};
\end{scope}
\begin{scope}[yshift=-2.8cm,xshift=6cm]
\node [anchor=west] (label1) at (0,0) {更多数据-总词数:100K $\sim$ 1M};
\node [anchor=north west] (p1) at (label1.south west) {$\textrm{P}('\textrm{很}')=0.000010$};
\node [anchor=north west] (p2) at (p1.south west) {$\textrm{P}('\textrm{。}')=0.001812$};
\node [anchor=north west] (p3) at (p2.south west) {$\textrm{P}('\textrm{确实}')=0.000001$};
\end{scope}
\node [anchor=north west] (label11) at ([xshift=18.0em,yshift=1.63em]label1.south west) {更多数据-总词数:100K $\sim$ 1M};
\node [anchor=north west] (p12) at (label11.south west) {$\textrm{P}('\textrm{很}')=0.000010$};
\node [anchor=north west] (p22) at (p12.south west) {$\textrm{P}('\textrm{。}')=0.001812$};
\node [anchor=north west] (p32) at (p22.south west) {$\textrm{P}('\textrm{确实}')=0.000001$};
\begin{pgfonlayer}{background}
{
\node[rectangle,draw=ublue, inner sep=0mm] [fit =(label1)(label11)(p3)(p32)] {};
}
\end{pgfonlayer}
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/perspectives-of-expert-ordinary-and-syntactic-parser.tex → Book/Chapter2/Figures/figure-perspectives-of-expert-ordinary-and-syntactic-parser.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
......@@ -115,7 +78,5 @@
\end{tabular}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/phrase-structure-tree-and-dependency-tree.tex → Book/Chapter2/Figures/figure-phrase-structure-tree-and-dependency-tree.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
......@@ -113,8 +79,4 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/probability-values-corresponding-to-different-derivations.tex → Book/Chapter2/Figures/figure-probability-values-corresponding-to-different-derivations.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
......@@ -120,7 +86,5 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/process-of-statistical-syntax-analysis.tex → Book/Chapter2/Figures/figure-process-of-statistical-syntax-analysis.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
......@@ -104,7 +69,6 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/rules-of-grammar.tex → Book/Chapter2/Figures/figure-rules-of-grammar.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
......@@ -76,7 +41,5 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-the-dice-game-model.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
......@@ -49,7 +13,7 @@
\node [draw,thick,minimum size=10pt] at (\i,0) {1};
}
}
\node [anchor=west] at (31em,0) {$\textrm{P}("1") = 5/30$};
\node [anchor=west] at (33em,0) {$\textrm{P}("1") = 5/30$};
\end{scope}
\begin{scope}[yshift=-2.5em]
......@@ -58,7 +22,7 @@
\node [draw,thick,minimum size=10pt] at (\i,0) {{\color{red} 2}};
}
}
\node [anchor=west] at (31em,0) {$\textrm{P}("2") = 4/30$};
\node [anchor=west] at (33em,0) {$\textrm{P}("2") = 4/30$};
\end{scope}
\begin{scope}[yshift=-5.0em]
......@@ -67,7 +31,7 @@
\node [draw,thick,minimum size=10pt] at (\i,0) {{\color{ublue} 3}};
}
}
\node [anchor=west] at (31em,0) {$\textrm{P}("3") = 6/30$};
\node [anchor=west] at (33em,0) {$\textrm{P}("3") = 6/30$};
\end{scope}
\begin{scope}[yshift=-7.5em]
......@@ -76,7 +40,7 @@
\node [draw,thick,minimum size=10pt] at (\i,0) {{\color{ugreen} 4}};
}
}
\node [anchor=west] at (31em,0) {$\textrm{P}("4") = 12/30$};
\node [anchor=west] at (33em,0) {$\textrm{P}("4") = 12/30$};
\end{scope}
\begin{scope}[yshift=-10.0em]
......@@ -85,7 +49,7 @@
\node [draw,thick,minimum size=10pt] at (\i,0) {{\color{purple} 5}};
}
}
\node [anchor=west] at (31em,0) {$\textrm{P}("5") = 2/30$};
\node [anchor=west] at (33em,0) {$\textrm{P}("5") = 2/30$};
\end{scope}
\begin{scope}[yshift=-12.5em]
......@@ -94,15 +58,11 @@
\node [draw,thick,minimum size=10pt] at (\i,0) {{\color{orange} 6}};
}
}
\node [anchor=west] at (31em,0) {$\textrm{P}("6") = 1/30$};
\node [anchor=west] at (33em,0) {$\textrm{P}("6") = 1/30$};
\end{scope}
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-the-dice-game.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
......@@ -88,10 +53,3 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-the-dice-game2.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
......@@ -88,7 +52,5 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/two-different-derivation-of-regulation.tex → Book/Chapter2/Figures/figure-two-different-derivation-of-regulation.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
......@@ -97,7 +59,5 @@
\end{minipage}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/usage-of-regulation.tex → Book/Chapter2/Figures/figure-usage-of-regulation.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
......@@ -77,7 +43,6 @@
\end{scope}
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/Figures/figure-word-segmentation-based-on-statistics.tex
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
......@@ -75,11 +40,11 @@
\end{pgfonlayer}
{
\draw [->,very thick,ublue] ([xshift=0.2em]corpus.east) -- ([xshift=3.2em]corpus.east) node [pos=0.5, above] {\color{red}{\scriptsize{统计学习}}};
\draw [->,very thick,ublue] ([xshift=0.2em]corpus.east) -- ([xshift=3.1em]corpus.east) node [pos=0.5, above] {\color{red}{\scriptsize{统计学习}}};
}
{
\draw [->,very thick,ublue] ([xshift=0.2em]model.east) -- ([xshift=3.2em]model.east) node [pos=0.5, above] {\color{red}{\scriptsize{穷举\&计算}}};
\draw [->,very thick,ublue] ([xshift=0.2em]model.east) -- ([xshift=3.1em]model.east) node [pos=0.5, above] {\color{red}{\scriptsize{穷举\&计算}}};
}
{\scriptsize
......@@ -88,7 +53,7 @@
\node [anchor=north west] (sent) at (sentlabel.south west) {\textbf{确实现在数据很多}};
}
{
\node [anchor=north west] (seg1) at ([xshift=0em]sent.south west) {确/实现/在/数/据很/多};
\node [anchor=north west] (seg1) at ([xshift=0.5em]sent.south west) {确/实现/在/数/据很/多};
\node [anchor=north west] (seg2) at (seg1.south west) {确实/现在/数据/很/多};
\node [anchor=north west] (seg3) at (seg2.south west) {确实/现在/数/据/很/多};
}
......@@ -114,7 +79,5 @@
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter2/chapter2.tex
......@@ -28,8 +28,8 @@
% 图2.1
\begin{figure}[htp]
\centering
\subfigure[机器翻译系统被看作一个黑盒] {\includegraphics[scale=1.0]{./Chapter2/Figures/figure-MT-system-as-a-black-box.pdf} }
\subfigure[机器翻系统 = 语言分析 + 翻译引擎] {\includegraphics[scale=1.0]{./Chapter2/Figures/figure-MT=language-analysis+translation-engine.pdf}}
\subfigure[机器翻译系统被看作一个黑盒] {\input{./Chapter2/Figures/figure-MT-system-as-a-black-box} }
\subfigure[机器翻系统 = 语言分析 + 翻译引擎] {\input{./Chapter2/Figures/figure-MT=language-analysis+translation-engine}}
\caption{机器翻译系统的结构对比}
\label{figureC2.1}
\end{figure}
......@@ -40,7 +40,7 @@
% 图2.2
\begin{figure}[htp]
\centering
\includegraphics[scale=1,width=3cm]{./Chapter2/Figures/figure-analysis-of-sentence-participle&syntactic.pdf}
\input{./Chapter2/Figures/figure-analysis-of-sentence-participle&syntactic}
\caption{中文句子“猫喜欢吃鱼”的分析结果（分词和句法分析）}
\label{figureC2.2}
\end{figure}
......@@ -113,7 +113,7 @@
\centering
\subfigure { \includegraphics[scale=1,height=4.5cm]{./Chapter2/Figures/figure-Probability-density-function.jpg} }
\subfigure {\includegraphics[scale=1,height=4.5cm]{./Chapter2/Figures/figure-Distribution-function.jpg}}
\caption{一个概率密度函数(左)与其对应的分布函数(右)}
\caption{{\red图没画}一个概率密度函数(左)与其对应的分布函数(右)}
\label{figureC2.3}
\end{figure}
%-------------------------------------------
......@@ -126,10 +126,10 @@
\parinterval 贝叶斯法则是条件概率计算时的重要依据，条件概率可以表示为
%----------------------------------------------
\begin{equation}
\begin{eqnarray}
\textrm{P}{(B|A)} = \frac{\textrm{P}(A\cap{B})}{\textrm{P}(A)} = \frac{\textrm{P}(A)\textrm{P}(B|A)}{\textrm{P}(A)} = \frac{\textrm{P}(B)\textrm{P}(A|B)}{\textrm{P}(A)}
\label{eqC2.1}
\end{equation}
\end{eqnarray}
%----------------------------------------------
\parinterval 边缘概率是和联合概率对应的，它指的是$\textrm{P}(X=a)$或$\textrm{P}(Y=b)$，即仅与单个随机变量有关的概率称为边缘概率。
......@@ -137,19 +137,19 @@
\parinterval 对于离散随机变量$X$和$Y$，我们知道$\textrm{P}(X,Y)$，则边缘概率$\textrm{P}(X)$可以通过求和的方式得到，如下式所示
%----------------------------------------------
\begin{equation}
\begin{eqnarray}
\forall x \in X ,\textrm{P}(X=x)=\sum_{y} \textrm{P}(X=x,Y=y)
\label{eqC2.2}
\end{equation}
\end{eqnarray}
%----------------------------------------------
\parinterval 对于连续变量，边缘概率$\textrm{P}(X)$需要通过积分得到，如下式所示
%----------------------------------------------
\begin{equation}
\begin{eqnarray}
\textrm{P}(X)=\int \textrm{P}(x,y)dy
\label{eqC2.3}
\end{equation}
\end{eqnarray}
%----------------------------------------------
\parinterval 为了更好的区分条件概率、边缘概率和联合概率，我们将通过图\ref{figureC2.4}所示的面积来举例说明。
......@@ -159,7 +159,7 @@
\begin{figure}[htp]
\centering
\includegraphics[scale=1,height=2cm]{./Chapter2/Figures/figure-schematic-edge-probability&joint-probability.png}
\caption{一个概率密度函数与其对应的分布函数}
\caption{{\red{图没画}}一个概率密度函数与其对应的分布函数}
\label{figureC2.4}
\end{figure}
%-------------------------------------------
......@@ -177,27 +177,21 @@
\parinterval 由条件概率公式$\textrm{P}(a \mid b)=\textrm{P}(ab)/\textrm{P}(b)$，我们知道条件概率的定义$\textrm{P}(a \mid b)$，其描述的是两个事件$a$和$b$，在$b$事件发生的条件下，$a$发生的概率。如果我们将其推广到三个事件$a$、$b$、$c$，为了计算$\textrm{P}(a,b,c)$我们可以运用两次$\textrm{P}(a \mid b)=\textrm{P}(ab)/\textrm{P}(b)$，计算过程如下：
%----------------------------------------------
\begin{equation}
\textrm{P}(a,b,c)=\textrm{P}(a \mid b ,c)\textrm{P}(b,c)
\begin{eqnarray}
\textrm{P}(a,b,c) & = & \textrm{P}(a \mid b ,c)\textrm{P}(b,c) \nonumber \\
\textrm{P}(b,c) & = & \textrm{P}(b \mid c)\textrm{P}(c)\nonumber \\
\textrm{P}(a,b,c) & = & \textrm{P}(a \mid b,c)\textrm{P}(b \mid c)\textrm{P}(c)
\label{eqC2.4}
\end{equation}
\begin{equation}
\textrm{P}(b,c)=\textrm{P}(b \mid c)\textrm{P}(c)
\label{eqC2.5}
\end{equation}
\begin{equation}
\textrm{P}(a,b,c)=\textrm{P}(a \mid b,c)\textrm{P}(b \mid c)\textrm{P}(c)
\label{eqC2.6}
\end{equation}
\end{eqnarray}
%----------------------------------------------
\parinterval 推广到$n$个事件，我们得到了链式法则的公式
%----------------------------------------------
\begin{equation}
\begin{eqnarray}
\textrm{P}(x_1,x_2,...,x_n)=\textrm{P}(x_1) \prod_{i=2}^n \textrm{P}(x_i \mid x_1,x_2,...,x_{(i-1)})
\label{eqC2.7}
\end{equation}
\label{eqC2.5}
\end{eqnarray}
%----------------------------------------------
\parinterval 我们可以通过下面这个例子更好的理解链式法则，如图所示，$A$、$B$、$C$、$D$、$E$分别代表五个事件，其中，$A$只和$B$有关，$C$只和$B$、$D$有关，$E$只和$C$有关，$B$和$D$不依赖其他任何事件。则$\textrm{P}(A,B,C,D,E)$的表达式如下式：
......@@ -208,25 +202,25 @@
\centering
\includegraphics[scale=1,height=2.5cm]{./Chapter2/Figures/figure-schematic-chain-rule.png}
\setlength{\belowcaptionskip}{-1cm}
\caption{A,B,C,D,E关系图}
\caption{{\red{图没画}}A,B,C,D,E关系图}
\label{figureC2.5}
\end{figure}
%-------------------------------------------
\begin{equation}
\begin{split}
\textrm{P}(A,B,C,D,E)&=\textrm{P}(E \mid A,B,C,D) \cdot \textrm{P}(A,B,C,D) \\
&=\textrm{P}(E \mid A,B,C,D) \cdot \textrm{P}(D \mid A,B,C) \cdot \textrm{P}(A,B,C) \\
&=\textrm{P}(E \mid A,B,C,D) \cdot \textrm{P}(D \mid A,B,C) \cdot \textrm{P}(C \mid A,B) \cdot \textrm{P}(A,B) \\
&=\textrm{P}(E \mid A,B,C,D) \cdot \textrm{P}(D \mid A,B,C) \cdot \textrm{P}(C \mid A,B) \cdot \textrm{P}(B \mid A) \cdot \textrm{P}(A)
\label{eqC2.8}
\end{split}
\end{equation}
\begin{eqnarray}
\textrm{P}(A,B,C,D,E)&=&\textrm{P}(E \mid A,B,C,D) \cdot \textrm{P}(A,B,C,D) \nonumber \\
&=&\textrm{P}(E \mid A,B,C,D) \cdot \textrm{P}(D \mid A,B,C) \cdot \textrm{P}(A,B,C) \nonumber \\
&=&\textrm{P}(E \mid A,B,C,D) \cdot \textrm{P}(D \mid A,B,C) \cdot \textrm{P}(C \mid A,B) \cdot \textrm{P}(A,B) \nonumber \\
&=&\textrm{P}(E \mid A,B,C,D) \cdot \textrm{P}(D \mid A,B,C) \cdot \textrm{P}(C \mid A,B) \cdot \textrm{P}(B \mid A) \cdot \textrm{P}(A)\nonumber \\
\label{eqC2.6}
\end{eqnarray}
\parinterval 根据图\ref {figureC2.5} 易知$E$只和$C$有关，所以$\textrm{P}(E \mid A,B,C,D)=\textrm{P}(E \mid C)$；$D$不依赖于其他事件，所以$\textrm{P}(D \mid A,B,C)=\textrm{P}(D)$；$C$只和$BD$有关，所以$\textrm{P}(C \mid A,B)=\textrm{P}(C \mid B)$；$B$不依赖于其他事件，所以$\textrm{P}(B \mid A)=\textrm{P}(B)$。最终化简可得：
\begin{equation}
%---------------------------------------------
\begin{eqnarray}
\textrm{P}(A,B,C,D,E)=\textrm{P}(E \mid C) \cdot \textrm{P}(D) \cdot \textrm{P}(C \mid B) \cdot \textrm{P}(B)
\label{eqC2.9}
\end{equation}
\label{eqC2.7}
\end{eqnarray}
%---------------------------------------------
\parinterval 由此可以看出使用链式法则可以大大减小求解概率表达式时的计算量。
......@@ -236,18 +230,18 @@
\parinterval 首先介绍一下全概率公式：全概率公式（Law of total probability）是概率论中重要的公式，它可以将一个复杂事件发生的概率分解成不同情况的小事件发生概率的和。这里我们先介绍一个概念——划分。
\parinterval 若集合S的一个划分事件为$B_1,…,B_n$是指它们满足
\begin{equation}
%---------------------------------------------
\begin{eqnarray}
\bigcup_{i=1}^n B_i=S \textrm{且}B_iB_j=\varnothing , i,j=1,...,n,i\neq j
\label{eqC2.9}
\end{equation}
\label{eqC2.8}
\end{eqnarray}
\parinterval 设$B_1,…,B_n$是S的一个划分，A为事件，则
\begin{equation}
\begin{eqnarray}
\textrm{P}(A)=\sum_{k=1}^n \textrm{P}(A \mid B_k)\textrm{P}(B_k)
\label{eqC2.9}
\end{equation}
\end{eqnarray}
%---------------------------------------------
\parinterval 这就是全概率公式。
\parinterval 举个例子，小张从家到公司有三条路分别为a，b，c，选择每条路的概率分别为0.5，0.3，0.2，那么：
......@@ -255,58 +249,58 @@
\parinterval $S_a$: 选择a路去上班，$S_b$: 选择b路去上班，$S_c$: 选择c路去上班 $S$：小张去上班,这四件事的关系即为：$S_a$，$S_b$，$S_c$是$S$的划分。
\parinterval 如果三条路不拥堵的概率分别为$\textrm{P}({S_{a}^{'}})$=0.2，$\textrm{P}({S_{b}^{'}})$=0.4，$\textrm{P}({S_{c}^{'}})$=0.7，那么事件L：小张上班没有遇到拥堵情况的概率就是：
\begin{equation}
\begin{split}
{\textrm{P}(L)} &= {\textrm{P}( L| S_a )\textrm{P}(S_a )+\textrm{P}( L| S_b )\textrm{P}(S_b )+\textrm{P}( L| S_c )\textrm{P}(S_c )}\\
& ={\textrm{P}({S_{a}^{'}})\textrm{P}(S_a)+\textrm{P}({S_{b}^{'}})\textrm{P}(S_b)+\textrm{P}({S_{c}^{'}})\textrm{P}(S_c) }\\
& ={0.36}\nonumber \\
\end{split}
\end{equation}
%--------------------------------------------
\begin{eqnarray}
{\textrm{P}(L)} &=& {\textrm{P}( L| S_a )\textrm{P}(S_a )+\textrm{P}( L| S_b )\textrm{P}(S_b )+\textrm{P}( L| S_c )\textrm{P}(S_c )}\nonumber \\
& = &{\textrm{P}({S_{a}^{'}})\textrm{P}(S_a)+\textrm{P}({S_{b}^{'}})\textrm{P}(S_b)+\textrm{P}({S_{c}^{'}})\textrm{P}(S_c) }\nonumber \\
& = &{0.36}\nonumber
\end{eqnarray}
%--------------------------------------------
%$\textrm{P}(L)=\textrm{P}( L| S_a )\textrm{P}(S_a )+\textrm{P}( L| S_b )\textrm{P}(S_b )+\textrm{P}( L| S_c )\textrm{P}(S_c )=\textrm{P}({S_{a}^{'}})\textrm{P}(S_a)+\textrm{P}({S_{b}^{'}})\textrm{P}(S_b)+\textrm{P}({S_{c}^{'}})\textrm{P}(S_c)=0.36$
\parinterval \textbf{贝叶斯法则（Bayes’ rule）}是概率论中的一个定理，通常用于知$\textrm{P}(A \mid B)$求$\textrm{P}(B \mid A)$。其内容如下：
%--------------------------------------------
\parinterval 设$B_1,…,B_n$是S的一个划分，A为事件，则对于$i=1,…,n$，有如下公式
\begin{equation}
\begin{eqnarray}
\textrm{P}(B_i \mid A)=\frac {\textrm{P}(A \mid B_i)\textrm{P}(B_i) } { \sum_{k=1}^n\textrm{P}(A \mid B_k)\textrm{P}(B_k) }
\label{eqC2.9}
\end{equation}
\label{eqC2.10}
\end{eqnarray}
\parinterval 下面我们来看看贝叶斯公式的推导。由前面的知识，我们知道条件概率的公式为
\begin{equation}
\begin{eqnarray}
\textrm{P}(B \mid A)= \frac {\textrm{P}(AB)} {\textrm{P}(A)}
\label{eqC2.9}
\end{equation}
\label{eqC2.11}
\end{eqnarray}
\parinterval 由乘法定理我们可以得到
\begin{equation}
\begin{eqnarray}
\textrm{P}(AB)=\textrm{P}(B)\textrm{P}(A \mid B)
\label{eqC2.9}
\end{equation}
\label{eqC2.12}
\end{eqnarray}
\parinterval 设$B_1,…,B_n$是S的一个划分，A为事件，由全概率公式我们可以得到
\begin{equation}
\begin{eqnarray}
\textrm{P}(A)=\textrm{P}(A \mid B_1)\textrm{P}(B_1)+\textrm{P}(A \mid B_2)\textrm{P}(B_2)+\ldots +\textrm{P}(A \mid B_n)\textrm{P}(B_n)
\label{eqC2.9}
\end{equation}
\label{eqC2.13}
\end{eqnarray}
\parinterval 将乘法定理带入条件概率的分子，将全概率公式带入条件概率的分母，我们就可以得到贝叶斯定理
\begin{equation}
\begin{eqnarray}
\textrm{P}(B_i \mid A)=\frac {\textrm{P}(A \mid B_i)\textrm{P}(B_i) } {\sum_{k=1}^n \textrm{P}(A\mid B_k)\textrm{P}(B_k)}
\label{eqC2.9}
\end{equation}
\label{eqC2.14}
\end{eqnarray}
\parinterval 由上式，我们也可以得到贝叶斯公式的另外两种写法
\begin{equation}
\begin{eqnarray}
\textrm{P}(A \mid B)=\frac { \textrm{P}(A \mid B)\textrm{P}(B) } {\textrm{P}(A)}
\label{eqC2.9}
\end{equation}
\begin{equation}
\label{eqC2.15}
\end{eqnarray}
\begin{eqnarray}
\textrm{P}(A \mid B)=\frac { \textrm{P}(A \mid B)\textrm{P}(B) } {\textrm{P}(A \mid B)\textrm{P}(B)+\textrm{P}(A \mid \bar{B}) \textrm{P}(\bar{B})}
\label{eqC2.9}
\end{equation}
\label{eqC2.16}
\end{eqnarray}
%--------------------------------------------
\parinterval 贝叶斯公式常用于根据已知的结果来推断使之发生的各因素的可能性。
......@@ -328,10 +322,10 @@
\parinterval 在这两句话中，“太阳从东方升起”是一件确定性事件，几乎不需要查阅更多信息就可以确认，因此这件事的信息熵相对较低；而“明天天气多云”这件事，我们需要询问气象局的相关研究人员，或者关注天气预报，才能大概率确定的一件事，它的不确定性很高，因而它的信息熵也就相对较高。因此，信息熵也是对事件不确定性的度量。
\parinterval 一个事件X的自信息（self-information）的表达式为：
\begin{equation}
\begin{eqnarray}
\textrm{I}(x)=-log\textrm{P}(x)
\label{eqC2.9}
\end{equation}
\label{eqC2.17}
\end{eqnarray}
\parinterval 其中，$\textrm{P}(x)$表示概率，自信息用来衡量单一事件发生时所包含的信息多少，当底数为e时，单位为$nats$，其中1$nats$是通过观察概率为$\frac{1}{e}$的事件而获得的信息量；当底数为2时，单位为$bits$或$shannons$，我们通常使用前者。$\textrm{I}(x)$和$\textrm{P}(x)$的函数关系如图\ref{figureC2.6} 所示：
......@@ -346,20 +340,20 @@
%-------------------------------------------
\parinterval 自信息只处理单一的结果。若量化整个概率分布中的不确定性或者说信息量，我们可以用信息熵，其公式如下：
\begin{equation}
\begin{eqnarray}
\textrm{H}(x)=\sum_{x \in \textrm{X}}[ \textrm{P}(x) \textrm{I}(x)] =- \sum_{x \in \textrm{X} } [\textrm{P}(x)log(\textrm{P}(x)) ]
\label{eqC2.9}
\end{equation}
\label{eqC2.18}
\end{eqnarray}
\parinterval 一个分布的信息熵也就是从该分布中得到的一个事件的期望信息量。比如，$a$、$b$、$c$、$d$三支球队，三支队伍夺冠的概率分别是$P1$、$P2$、$P3$、$P4$，某个人对比赛不感兴趣但是又想知道哪只球队夺冠，通过使用二分法2次就确定哪支球队夺冠了。但其实，我们知道这四只球队中c的实力比较强劲，那么猜1次就可以确定。所以对于前者，哪只球队夺冠的信息量较高，信息熵也相对较高，对于后者信息量和信息熵也就相对较低。因此我们可以得知：较为尖锐的分布具有较低的熵；分布越接近均匀熵越大。
\subsubsection{（二）KL距离}\index{Chapter2.2.5.2}
\parinterval 如果在相同的随机变量$X$上有两个独立的概率分布P$(x)$和Q$(x)$，那么我们可以使用KL距离("Kullback-Leibler" 散度)来衡量这两个分布的不同，也就是大家所说的相对熵，其公式如下：
\begin{equation}
\begin{eqnarray}
\textrm{D}_{KL}(P\parallel Q)=\sum_{x \in \textrm{X}} [ \textrm{P}(x)log \frac{\textrm{P}(x) }{ \textrm{Q}(x) } ] =\sum_{x \in \textrm{X} }[ \textrm{P}(x)(log\textrm{P}(x)-log \textrm{Q}(x))]
\label{eqC2.9}
\end{equation}
\label{eqC2.19}
\end{eqnarray}
\parinterval 这一概念的意义在于：在相同事件空间里，概率分布$\textrm{P}(x)$对应的每个事件，若用概率分布Q$(x)$编码时，平均每个基本事件的信息量增加了多少。它衡量的是相同事件空间里的两个概率分布的差异情况。KL距离有两条重要的性质：
......@@ -374,10 +368,10 @@
\subsubsection{（三）交叉熵}\index{Chapter2.2.5.3}
\parinterval 交叉熵是一个与KL距离密切相关的概念，它的公式是：
\begin{equation}
\begin{eqnarray}
\textrm{H}(P,Q)=-\sum_{x \in \textrm{X}} [\textrm{P}(x) log \textrm{Q}(x) ]
\label{eqC2.9}
\end{equation}
\label{eqC2.20}
\end{eqnarray}
\parinterval 结合相对熵公式可知，交叉熵是KL距离公式中的右半部分。因此，求关于Q的交叉熵的最小值等价于求KL距离的最小值。交叉熵与KL距离的意义相同：都是用来描述两个分布的差异，由于交叉熵计算上更加直观方便，因此在机器翻译中被广泛应用。
......@@ -386,10 +380,10 @@
\parinterval 对于机器翻译系统而言，输入的是已经切分好的单词序列，而不是原始的字符串。比如，对于一个中文句子，单词之间是没有间隔的，因此我们需要把一个个的单词切分出来，这样机器翻译系统可以区分不同单元。甚至，我们可以对语言学上的单词进行进一步切分，得到词片段序列（比如：中国人=>中国 人）。我们可以把上述过程看作是一种分词过程，即：将一个输入的自然语言字符串切割成单元序列（token序列），每个单元都对应可以处理的最小单位。
%----------------------------------------------
% 图2.6
% 图2.7
\begin{figure}[htp]
\centering
\includegraphics[scale=1]{./Chapter2/Figures/figure-a-simple-pre-processing-process.pdf}
\input{./Chapter2/Figures/figure-a-simple-pre-processing-process}
\caption{一个简单的预处理流程}
\label{figureC2.7}
\end{figure}
......@@ -450,10 +444,10 @@
\parinterval 在使用基于词典的分词方法时，只需预先加载词典到计算机中，扫描输入句子，查询每个词串是否出现在词典中。如图\ref{figureC2.8} 所示，比如，我们有一个包含六个词的词典，给定输入句子“确实现在物价很高”后，我们自左至右遍历输入句子的每个字，发现词串“确实”在词典中出现，说明“确实”是一个“词”，进行分词操作并在切分该“词”之后重复这个过程。
%----------------------------------------------
% 图2.7
% 图2.8
\begin{figure}[htp]
\centering
\includegraphics[scale=1]{./Chapter2/Figures/figure-Example-of-word-segmentation-based-on-dictionary.pdf}
\input{./Chapter2/Figures/figure-Example-of-word-segmentation-based-on-dictionary}
\setlength{\belowcaptionskip}{-0.2cm}
\caption{基于词典进行分词实例}
\label{figureC2.8}
......@@ -463,10 +457,10 @@
\parinterval 但是，基于词典的分词方法很“硬”。这是因为自然语言非常灵活，经常出现歧义，用词典定义的合法单词之间有重叠的交叉型歧义就很难解决。图\ref{figureC2.9} 就给出了上面例子中的交叉型歧义，从词典中查看，“实现”和“现在”都是合法的单词，但是在句子中二者有重叠，因此词典无法告诉我们哪个结果是正确的。
%----------------------------------------------
% 图2.8
% 图2.9
\begin{figure}[htp]
\centering
\includegraphics[scale=1]{./Chapter2/Figures/figure-cross-type-word-segmentation-ambiguity.pdf}
\input{./Chapter2/Figures/figure-cross-type-word-segmentation-ambiguity}
\setlength{\belowcaptionskip}{-0.2cm}
\caption{交叉型分词歧义}
\label{figureC2.9}
......@@ -495,10 +489,10 @@
%----------------------------------------------
% 图2.9
% 图2.10
\begin{figure}[htp]
\centering
\includegraphics[scale=1]{./Chapter2/Figures/figure-word-segmentation-based-on-statistics.pdf}
\input{./Chapter2/Figures/figure-word-segmentation-based-on-statistics}
\caption{基于统计的分词流程}
\label{figureC2.10}
\end{figure}
......@@ -514,36 +508,34 @@
% 图2.11
\begin{figure}[htp]
\centering
\includegraphics[scale=1]{./Chapter2/Figures/figure-the-dice-game.pdf}
\input{./Chapter2/Figures/figure-the-dice-game}
\setlength{\belowcaptionskip}{-0.2cm}
\caption{骰子结果}
\label{figureC2.11}
\end{figure}
%-------------------------------------------
\parinterval 似乎玩家的胜利只能来源于运气。不过，请注意，这里的假设“随便选一个数字”这本身就是一个概率模型，它对骰子的六个面的出现做了均匀分布假设。
\begin{equation}
\begin{eqnarray}
\textrm{P}("1")=\textrm{P}("2")=...=\textrm{P}("5")=\textrm{P}("6")=1/6
\label{eqC2.9}
\end{equation}
\label{eqC2.21}
\end{eqnarray}
\parinterval 但是这个游戏没有人规定骰子是均匀的（有些被坑了的感觉）。但是如果骰子的六个面不均匀呢？我们可以用一种更加“聪明”的方式定义一个新模型，即定义骰子的每一个面都以一定的概率出现，而不是相同的概率。这里，为了保证概率的归一性，我们只需定义$\theta_1 \sim \theta_5$，最后一个面的概率用1减去前几个面的概率之和进行表示，即
\begin{equation}
\begin{split}
\textrm{P}("1") &=\theta_1 \\
\textrm{P}("2") &=\theta_2 \\
\textrm{P}("3") &=\theta_3 \\
\textrm{P}("4") &=\theta_4 \\
\textrm{P}("5") &=\theta_5 \\
\textrm{P}("6") &=1-\sum_{1 \leq i \leq 5}\theta_i \qquad \lhd \textrm {归一性}
\label{eqC2.9}
\end{split}
\end{equation}
\begin{eqnarray}
\textrm{P}("1") &=&\theta_1 \nonumber \\
\textrm{P}("2") &=&\theta_2 \nonumber \\
\textrm{P}("3") &=&\theta_3 \nonumber \\
\textrm{P}("4") &=&\theta_4 \nonumber \\
\textrm{P}("5") &=&\theta_5 \nonumber \\
\textrm{P}("6") &=&1-\sum_{1 \leq i \leq 5}\theta_i \qquad \lhd \textrm {归一性}
\label{eqC2.22}
\end{eqnarray}
\parinterval 这里$\theta_1 \sim \theta_5$可以被看作是模型的参数。对于这样的模型，参数确定了，模型也就确定了。但是，新的问题来了，在定义骰子每个面的概率后，如何求出具体的值呢？一种常用的方法是，从大量实例中学习模型参数，这个方法也是常说的参数估计。我们可以将这个不均匀的骰子先实验性的掷很多次，这可以被看作是独立同分布的若干次采样，比如X次，发现“1”出现X1次，“2”出现X2次，以此类推，得到了各个面出现的次数。假设掷骰子中每个面出现的概率符合多项式分布，通过简单的概率论知识可以知道每个面出现概率的极大似然估计为：
\begin{equation}
\begin{eqnarray}
\textrm{P}("i")=\frac {X_i}{X}
\label{eqC2.9}
\end{equation}
\label{eqC2.23}
\end{eqnarray}
\parinterval 当X足够大的话，$\frac{X_i}{X}$可以无限逼近P(“i”)的真实值，因此可以通过大量的实验推算出掷骰子各个面的概率的准确估计值。回归到我们的问题中，如果我们在正式开始游戏前，预先掷骰子30次，得到如图\ref{figureC2.12}的结果。
......@@ -551,7 +543,7 @@
% 图2.12
\begin{figure}[htp]
\centering
\includegraphics[scale=1]{./Chapter2/Figures/figure-the-dice-game2.pdf}
\input{./Chapter2/Figures/figure-the-dice-game2}
\caption{预投骰子结果}
\label{figureC2.12}
\end{figure}
......@@ -563,7 +555,7 @@
% 图2.13
\begin{figure}[htp]
\centering
\includegraphics[scale=1]{./Chapter2/Figures/figure-the-dice-game-model.pdf}
\input{./Chapter2/Figures/figure-the-dice-game-model}
\caption{预设的骰子模型}
\label{figureC2.13}
\end{figure}
......@@ -591,7 +583,7 @@
% 图2.14
\begin{figure}[htp]
\centering
\includegraphics[scale=1]{./Chapter2/Figures/figure-full-probability-word-segmentation-1.pdf}
\input{./Chapter2/Figures/figure-full-probability-word-segmentation-1}
\caption{新投骰子结果}
\label{figureC2.14}
\end{figure}
......@@ -613,7 +605,7 @@
% 图2.15
\begin{figure}[htp]
\centering
\includegraphics[scale=1]{./Chapter2/Figures/figure-full-probability-word-segmentation-2.pdf}
\input{./Chapter2/Figures/figure-full-probability-word-segmentation-2}
\setlength{\belowcaptionskip}{-0.2cm}
\caption{换成汉字后结果}
\label{figureC2.15}
......@@ -625,7 +617,7 @@
% 图2.16
\begin{figure}[htp]
\centering
\includegraphics[scale=1,height=2.5cm]{./Chapter2/Figures/figure-full-probability-word-segmentation-3.pdf}
\input{./Chapter2/Figures/figure-full-probability-word-segmentation-3}
\caption{每个单词概率估计值}
\label{figureC2.16}
\end{figure}
......@@ -639,20 +631,18 @@
\parinterval 这里可以使用“大题小做”的技巧：原始的问题很复杂，我们将其切分为小问题。这样，将复杂的分词问题简单化，基于独立性假设解决分词问题：假定所有词出现都是相互独立的。设$w_1 w_2 w_3…w_m$表示一个由单词$w_1,w_2,w_3,…,w_m$组成的切分结果，于是有：
%-------------------------------------------
\begin{equation}
\begin{eqnarray}
\textrm{P}(w_1 w_2 w_3…w_m)=\textrm{P}(w_1) \cdot \textrm{P}(w_2) \cdot ... \cdot \textrm{P}(w_m)
\label{eqC2.9}
\end{equation}
\label{eqC2.25 }
\end{eqnarray}
\begin{equation}
\begin{split}
\textrm{P}\textrm{(‘确实/现在/数据/很/多’)} \\
& = \textrm{P}\textrm{'确实'} \cdot \textrm{P}\textrm{'现在'} \cdot \textrm{P}\textrm{'数据'} \cdot \textrm{P}\textrm{'很'} \cdot \textrm{P}\textrm{'多'} \\
& = 0.000001 \times 0.000022 \times 0.000009 \times 0.000010 \times 0.000078 \\
& = 1.5444 \times 10^{-25}
\label{eqC2.9}
\end{split}
\end{equation}
\begin{eqnarray}
&\textrm{P}&\textrm{(‘确实/现在/数据/很/多’)} \nonumber \\
& = &\textrm{P}\textrm{'确实'} \cdot \textrm{P}\textrm{'现在'} \cdot \textrm{P}\textrm{'数据'} \cdot \textrm{P}\textrm{'很'} \cdot \textrm{P}\textrm{'多'} \nonumber \\
& = &0.000001 \times 0.000022 \times 0.000009 \times 0.000010 \times 0.000078 \nonumber \\
& = &1.5444 \times 10^{-25}
\label{eqC2.26}
\end{eqnarray}
\parinterval 以“确实现在数据很多”这个实例来说，如果把这句话按照“确实/现在/数据/很/多”这样的方式进行切分，这句切分的概率P(“确实/现在/数据/很/多”)可以通过每个词出现概率相乘的方式进行计算。这个假设也是自然语言处理中1-gram语言模型假设，即当前词的生成与任何历史都无关。当然，独立性假设并不能完美描述客观世界的问题，但是它大大化简了问题的复杂度。
......@@ -660,7 +650,7 @@
% 图2.17
\begin{figure}[htp]
\centering
\includegraphics[scale=1]{./Chapter2/Figures/figure-examples-of-Chinese-word-segmentation-based-on-1-gram-model.pdf}
\input{./Chapter2/Figures/figure-examples-of-Chinese-word-segmentation-based-on-1-gram-model}
\caption{基于1-gram语言模型的中文分词实例}
\label{figureC2.17}
\end{figure}
......@@ -693,19 +683,19 @@
\parinterval 直接求$\textrm{P}(w_1 w_2...w_m)$并不简单，因为如果把$w_1 w_2...w_m$整个作为一个变量，模型的参数量会非常大。$w_1 w_2...w_m$有$|V|^m$种可能性，这里$|V|$表示词汇表大小。显然，当$m$增大的时候会使模型复杂度会急剧增加，甚至都无法进行存储和计算。既然把$w_1 w_2...w_m$整个作为一个变量不好处理，就可以考虑对这个序列的生成进行分解。使用链式法则，很容易得到
\begin{equation}
\begin{eqnarray}
\textrm{P}(w_1 w_2...w_m)=\textrm{P}(w_1)\textrm{P}(w_2|w_1)\textrm{P}(w_3|w_1 w_2)...\textrm{P}(w_m|w_1 w_2...w_{m-1})
\label{eq:2.4.1.1}
\end{equation}
\end{eqnarray}
这样，$w_1 w_2...w_m$的生成可以被看作是逐个生成每个单词的过程，即首先生成$w_1$，然后根据$w_1$再生成$w_2$，然后根据$w_1 w_2$再生成$w_3$，以此类推，直到根据所有前$m$-1个词生成序列的最后一个单词$w_m$。这个模型把联合概率$\textrm{P}(w_1 w_2...w_m)$分解为多个条件概率的乘积，虽然可以对生成序列的过程进行分解，但是模型的复杂度和以前是一样的，比如，$\textrm{P}(w_m|w_1 w_2...w_{m-1})$仍然不好计算。
\parinterval 换一个角度看，$\textrm{P}(w_m|w_1 w_2...w_{m-1})$体现了一种基于“历史”的单词生成模型，也就是把前面生成的所有单词作为“历史”，并参考这个“历史”生成当前单词。但是这个“历史”的长度和整个序列长度是相关的，也是一种长度变化的历史序列。为了化简问题，一种自然的想法是使用定长历史，比如，每次只考虑前面$n$-1个历史单词来生成当前单词，这就是$n$-gram语言模型。这个模型的数学描述如下：
\begin{equation}
\begin{eqnarray}
\textrm{P}(w_m|w_{m-n+1}...w_{m-1})=\textrm{P}(w_m|w_1 w_2...w_{m-1})
\label{eq:2.4.1.2}
\end{equation}
\end{eqnarray}
\parinterval 这样，整个序列$w_1 w_2...w_m$的生成概率可以被重新定义为：
%------------------------------------------------------
......@@ -740,10 +730,10 @@ $\textrm{P}(w_m|w_1 ... w_{m-1})$ & $\textrm{P}(w_m)$ & $\textrm{P}(w_m|w_{m-1})
\begin{itemize}
\item 极大似然估计。直接利用不同词序列在训练数据中出现的频度计算出$\textrm{P}(w_m|w_{m-n+1} ... w_{m-1})$
\begin{equation}
\begin{eqnarray}
\textrm{P}(w_m|w_{m-n+1}...w_{m-1})=\frac{count(w_{m-n+1}...w_m)}{count(w_{m-n+1}...w_{m-1})}
\label{eq:2.4.1.3}
\end{equation}
\end{eqnarray}
\item 人工神经网络方法。构建一个人工神经网络估计$\textrm{P}(w_m|w_{m-n+1} ... w_{m-1})$的值，比如，可以构建一个前馈神经网络来对$n$-gram进行建模。
\end{itemize}
......@@ -753,13 +743,12 @@ $\textrm{P}(w_m|w_1 ... w_{m-1})$ & $\textrm{P}(w_m)$ & $\textrm{P}(w_m|w_{m-1})
\parinterval 使用$n$-gram语言模型非常简单。我们可以像2.3.2节中一样，直接用它来对词序列出现的概率进行计算。比如，可以使用一个2-gram语言模型计算一个分词序列的概率
\begin{equation}
\begin{aligned}
& \textrm{P}_{2-gram}{('\textrm{确实}/\textrm{现在}/\textrm{数据}/\textrm{很}/\textrm{多}')} \\ = \quad & \textrm{P}('\textrm{确实}')\times\textrm{P}('\textrm{现在}'|'\textrm{确实}')\times\textrm{P}('\textrm{数据}'|'\textrm{现在}')\times \\
& \textrm{P}('\textrm{很}'|'\textrm{数据}')\times\textrm{P}('\textrm{多}'|'\textrm{很}')
\begin{eqnarray}
& &\textrm{P}_{2-gram}{('\textrm{确实}/\textrm{现在}/\textrm{数据}/\textrm{很}/\textrm{多}')} \nonumber \\
&= & \textrm{P}('\textrm{确实}') \times\textrm{P}('\textrm{现在}'|'\textrm{确实}')\times\textrm{P}('\textrm{数据}'|'\textrm{现在}') \nonumber \\
& \times & \textrm{P}('\textrm{很}'|'\textrm{数据}')\times\textrm{P}('\textrm{多}'|'\textrm{很}')
\label{eq:2.4.1.4}
\end{aligned}
\end{equation}
\end{eqnarray}
\parinterval 以$n$-gram语言模型为代表的统计语言模型的应用非常广泛。除了分词，在文本生成、信息检索、摘要等等自然语言处理任务中，语言模型都有举足轻重的地位。包括近些年非常受关注的预训练模型，本质上也是统计语言模型。这些技术都会在后续章节进行介绍。值得注意的是，统计语言模型给我们解决自然语言处理问题提供了一个非常好的建模思路，即：把整个序列生成的问题转化为逐个生成单词的问题。很快我们就会看到，这种建模方式会被广泛的用于机器翻译建模中，在统计机器翻译和神经机器翻译中都会有明显的体现。
......@@ -768,10 +757,10 @@ $\textrm{P}(w_m|w_1 ... w_{m-1})$ & $\textrm{P}(w_m)$ & $\textrm{P}(w_m|w_{m-1})
\parinterval 在式\ref{eq:2.4.1.4}的例子中，如果语料中从没有“确实”和“现在”两个词连续出现的情况，那么使用2-gram计算 “确实/现在/数据/很/多”的切分方式的概率时，会出现如下情况
\begin{equation}
\begin{eqnarray}
\textrm{P}('\textrm{现在}'|'\textrm{确实}') = \frac{count('\textrm{确实}\,\textrm{现在}')}{count('\textrm{确实}')} = \frac{0}{count('\textrm{确实}')} = 0
\label{eq:2.4.1.5}
\end{equation}
\end{eqnarray}
\parinterval 显然，这个结果是不能接受的。因为即使语料中没有 “确实”和“现在”两个词连续出现，但是这种搭配也是客观存在的。这时简单的用极大似然估计得到概率却是0，导致整个切分结果的概率为0。更常见的问题是那些根本没有出现在词表中的词，称为未登录词（Out-of-Vocabulary, OOV），比如一些生僻词，可能模型训练阶段从来没有看到过，这时模型仍然会给出0概率。图\ref{fig:2.4.1.1}展示了词语出现频度的分布，可以看到绝大多数词都是低频词。
......@@ -797,33 +786,34 @@ $\textrm{P}(w_m|w_1 ... w_{m-1})$ & $\textrm{P}(w_m)$ & $\textrm{P}(w_m|w_{m-1})
\parinterval 通常情况下，我们会利用采集到的语料库来模拟现实生活中真实全部的语料库。但是采集总是不充分的，比如无法涵盖所有的词汇，直接依据这样语料所获得的统计信息计算现实中的语言概率就会产生偏差。假设依据某语料C（从未出现“确实 现在”二元语法），评估一个已经分好词的句子S =“确实 现在 物价 很 高”的概率，当计算“确实 现在”的概率时使得评估$\textrm{P}(S) = 0$。显然这个结果是不够准确的，根据我们的常识，句子$S$是有出现的可能性的，这样句子的概率值不应该是0。
\begin{equation}
\begin{eqnarray}
\textrm{P}(\textrm{现在}|\textrm{确实}) = \frac{count(\textrm{确实}\,\textrm{现在})}{count(\textrm{确实})} = \frac{0}{count(\textrm{确实})} = 0
\label{eq:2.4.1.6}
\end{equation}
\end{eqnarray}
\parinterval 为了避免这种由于数据所产生的评估预测概率为0的问题，采用“数据平滑”的方式对最大似然估计方法进行调整。通常的平滑方法都是为了提高低概率（如零概率），或者降低高概率，这种做法的思想比较类似于“劫富济贫”。
\parinterval 加法平滑方法（additive smoothing）假设每个n元语法出现的次数比实际统计次数多$\theta$次，$0 \leqslant\theta\leqslant 1$，使得分子部分不为0，那么计算前文例子“确实 现在”的概率时，可以使用如下方法计算。
\begin{equation}
\begin{eqnarray}
\textrm{P}(\textrm{现在}|\textrm{确实}) = \frac{\theta + count(\textrm{确实}\,\textrm{现在})}{\sum_{w}^{|V|}(\theta + count(\textrm{确实}w))} = \frac{\theta + count(\textrm{确实}\,\textrm{现在})}{\theta{|V|} + count(\textrm{确实})}
\label{eq:2.4.1.7}
\end{equation}
\end{eqnarray}
\parinterval 这里面$V$表示所有词汇的词表，$|V|$为词表中单词的个数，$w$为词典中的词。常见的加法平滑方法会将$\theta$取1，这时我们又称为加一平滑或是拉普拉斯平滑。这种方法比较容易理解，也比较简单，但是一些人认为这种方法的表现较差，因此，其实际的使用效果还要视具体情况而定。
\parinterval 举一个例子来形象的描述加法平滑方法。假设在一个英文文档中随机抽取词汇，已经抽到的词包括12个，词典大小$|V|$=20，已抽到的词汇统计结果为：4 look，3 people，2 am，1 what，1 want，1 do。为了更形象的描述在平滑之前和平滑之后的概率分布的区别，如图所示：
%----------------------------------------------
% 图2.4.1.2
% 图2.19
\begin{figure}[htp]
\centering
\includegraphics{./Chapter2/Figures/figure-no-smoothing&smoothed-probability-distributions.jpg}
\caption{无平滑和有平滑后的概率分布（{\red需要高清图片}）}
\subfigure[无平滑的概率分布] {\includegraphics[scale=0.25]{./Chapter2/Figures/figure-no-smoothing&smoothed-probability-distributions-1.pdf} }
\subfigure[加一平滑的概率分布] {\includegraphics[scale=0.25]{./Chapter2/Figures/figure-no-smoothing&smoothed-probability-distributions-2.pdf}}
\caption{无平滑和有平滑后的概率分布}
\label{fig:2.4.1.2}
\end{figure}
%---------------------------
%-------------------------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\subsubsection{古德-图灵估计法}\index{Chapter2.4.2.2}
......@@ -832,31 +822,31 @@ $\textrm{P}(w_m|w_1 ... w_{m-1})$ & $\textrm{P}(w_m)$ & $\textrm{P}(w_m|w_{m-1})
\parinterval 假定在语料库中出现r次的$n$元语法有$n_r$个，特别的，出现0次的$n$元语法(即未登录词)出现的次数为$n_0$个。语料库中全部词语的个数为$N$，显然
\begin{equation}
\begin{eqnarray}
N = \sum_{r=1}^{\infty}{r\,n_r}
\label{eq:2.4.1.8}
\end{equation}
\end{eqnarray}
\parinterval 这时，出现$r$次的$n$元语法在词典中的相对频率为$r/N$，这也是不做平滑处理时这些词的概率估计。为了解决零概率问题，Good-Turing方法对于任何一个出现$r$次的$n$元语法，利用出现$r$+1次的$n$元语法统计量重新假设它出现$r^*$次，这里
\begin{equation}
\begin{eqnarray}
r^* = (r + 1)\frac{n_{r + 1}}{n_r}
\label{eq:2.4.1.9}
\end{equation}
\end{eqnarray}
\parinterval 基于这个公式，就可以估计所有0次$n$元语法的频次$n_0 r_0^*=(r_0+1)n_1=n_1$。要把这个重新估计的统计数转化为概率，只需要进行归一化处理：对于每个统计数为$r$的事件，其概率为$\textrm{p}_r=r^*/N$，其中
\begin{equation}
\begin{eqnarray}
N = \sum_{r=0}^{\infty}{r^{*}n_r} = \sum_{r=0}^{\infty}{(r + 1)n_{r + 1}} = \sum_{r=1}^{\infty}{r\,n_r}
\label{eq:2.4.1.10}
\end{equation}
\end{eqnarray}
也就是说，N仍然为这个整个样本分布最初的计数。这样样本中所有事件的概率之和为：
\begin{equation}
\begin{eqnarray}
N = \sum_{r>0}{p_r n_r} = 1 - \frac{n_1}{N} < 1
\label{eq:2.4.1.11}
\end{equation}
\end{eqnarray}
其中$n_1/N$的概率余量就是分配给所有统计为0的事件。
......@@ -890,10 +880,10 @@ N = \sum_{r>0}{p_r n_r} = 1 - \frac{n_1}{N} < 1
\parinterval 首先介绍一下absolute discounting平滑算法，公式如下所示
\begin{equation}
\begin{eqnarray}
\textrm{P}_{\textrm{AbsDiscount}}(w_i | w_{i-1}) = \frac{c(w_{i-1},w_i )-d}{c(w_{i-1})} + \lambda(w_{i-1})\textrm{P}(w)
\label{eq:2.4.1.12}
\end{equation}
\end{eqnarray}
其中$d$是固定的被裁剪的值，$\lambda$是一个正则化常数。可以看到第一项是经过减值调整过的2-gram的概率值，第二项则相当于一个带权重$\lambda$的1-gram的插值项。然而这种插值模型极易受到原始1-gram模型的干扰。
......@@ -903,31 +893,31 @@ N = \sum_{r>0}{p_r n_r} = 1 - \frac{n_1}{N} < 1
\parinterval 为了评估$\textrm{P}_{cont}$，统计使用当前词作为第二个词所出现二元语法的种类，二元语法种类越多，这个词作为第二个词出现的可能性越高，呈正比：
\begin{equation}
\begin{eqnarray}
\textrm{P}_{\textrm{cont}}(w_i) \varpropto |w_{i-1}: c(w_{i-1} w_i )>0|
\label{eq:2.4.1.13}
\end{equation}
\end{eqnarray}
通过全部的二元语法的种类做归一化可得到评估的公式
\begin{equation}
\begin{eqnarray}
\textrm{P}_{\textrm{cont}}(w_i) = \frac{|\{ w_{i-1}:c(w_{i-1} w_i )>0 \}|}{|\{ (w_{j-1}, w_j):c(w_{j-1},w_j )>0 \}|}
\label{eq:2.4.1.14}
\end{equation}
\end{eqnarray}
\parinterval 基于分母的变化还有另一种形式
\begin{equation}
\begin{eqnarray}
\textrm{P}_{\textrm{cont}}(w_i) = \frac{|\{ w_{i-1}:c(w_{i-1} w_i )>0 \}|}{\sum_{w^{\prime}}|\{ w_{i-1}^{\prime}:c(w_{i-1}^{\prime},w_i^{\prime} )>0 \}|}
\label{eq:2.4.1.15}
\end{equation}
\end{eqnarray}
结合基础的absolute discounting计算公式，从而得到了Kneser-Ney平滑方法的公式
\begin{equation}
\begin{eqnarray}
\textrm{P}_{\textrm{KN}}(w_i|w_{i-1}) = \frac{\max(c(w_{i-1},w_i )-d,0)}{c(w_{i-1})}+ \lambda(w_{i-1})\textrm{P}_{\textrm{cont}}(w_i)
\label{eq:2.4.1.16}
\end{equation}
\end{eqnarray}
\noindent 其中
\begin{eqnarray}
......@@ -972,7 +962,7 @@ c_{\textrm{KN}}(\cdot) & = & \begin{cases} \textrm{count}(\cdot)\quad for\ the\
% 图2.5.1.1
\begin{figure}[htp]
\centering
\includegraphics{./Chapter2/Figures/phrase-structure-tree-and-dependency-tree.pdf}
\input{./Chapter2/Figures/figure-phrase-structure-tree-and-dependency-tree}
\caption{短语结构树(左)和依存树(右)}
\label{fig:phrase-structure-tree-and-dependency-tree}
\end{figure}
......@@ -1037,28 +1027,28 @@ c_{\textrm{KN}}(\cdot) & = & \begin{cases} \textrm{count}(\cdot)\quad for\ the\
\parinterval 举例说明，假设有上下文无关文法$G=<N,\Sigma,R,S>$，其中把非终结符集合定义为不同的句法标记
\begin{equation}
\begin{eqnarray}
\textrm{N}=\{\textrm{NN},\textrm{VV},\textrm{NP},\textrm{VP},\textrm{IP}\}
\end{equation}
\end{eqnarray}
这里，\textrm{NN}代表名词，\textrm{VV}代表动词，\textrm{NP}代表名词短语，\textrm{VP}代表动词短语，\textrm{IP}代表单句。进一步，把终结符集合定义为
\begin{equation}
\begin{eqnarray}
\Sigma = \{\text{猫,喜欢,吃,鱼}\}
\end{equation}
\end{eqnarray}
再定义起始符集合为
\begin{equation}
\begin{eqnarray}
\textrm{S}=\{\textrm{IP}\}
\end{equation}
\end{eqnarray}
最后，文法的规则集定义如下（其中$r_i$为规则的编号）
%----------------------------------------------
% 图2.5.2.1
\begin{figure}[htp]
\centering
\includegraphics{./Chapter2/Figures/rules-of-grammar.pdf}
\input{./Chapter2/Figures/figure-rules-of-grammar}
\caption{还没有写}
\label{fig:2.5.2.1}
\end{figure}
......@@ -1076,7 +1066,7 @@ c_{\textrm{KN}}(\cdot) & = & \begin{cases} \textrm{count}(\cdot)\quad for\ the\
\begin{figure}[htp]
\centering
\includegraphics{./Chapter2/Figures/usage-of-regulation.pdf}
\input{./Chapter2/Figures/figure-usage-of-regulation}
\caption{规则的使用示意图}
\label{fig:2.5.2.2}
\end{figure}
......@@ -1111,20 +1101,20 @@ s_0 \overset{r_1}{\Rightarrow} s_1 \overset{r_2}{\Rightarrow} s_2 \overset{r_3}{
% 图2.5.2.3
\begin{figure}[htp]
\centering
\includegraphics{./Chapter2/Figures/example-of-derivation.pdf}
\input{./Chapter2/Figures/figure-example-of-derivation}
\caption{上下文无关文法推导实例}
\label{fig:2.5.2.3}
\end{figure}
%-------------------------------------------
\parinterval 通常，可以把推导简记为$\textrm{d}=r_1，r_2，…，r_n$，其中$∘$表示规则的组合。显然，$\textrm{d}$也对应了树形结构，也就是句法分析结果。从这个角度看，推导就是描述句法分析树的一种方式。此外，规则的推导也把规则的使用和生成的字符串对应起来。一个推导所生成的字符串，也被称作文法的一个句子。而一个文法所能生成的所有句子是这个文法所对应的语言。
\parinterval 通常，可以把推导简记为$\textrm{d}=r_1,r_2,…,r_n$，其中$ \cdot $表示规则的组合。显然，$\textrm{d}$也对应了树形结构，也就是句法分析结果。从这个角度看，推导就是描述句法分析树的一种方式。此外，规则的推导也把规则的使用和生成的字符串对应起来。一个推导所生成的字符串，也被称作文法的一个句子。而一个文法所能生成的所有句子是这个文法所对应的语言。
\parinterval 但是，句子和规则的推导并不是一一对应的。同一个句子，往往有很多推导与之对应，我们称为歧义。甚至同一棵句法树，也可以对应不同的推导。图\ref{fig:2.5.2.4}给出一个同一棵句法树所对应的两种不同的规则推导。
%-------------------------------------------
%图2.5.2.4
\begin{figure}[htp]
\centering
\includegraphics{./Chapter2/Figures/two-different-derivation-of-regulation.pdf}
\input{./Chapter2/Figures/figure-two-different-derivation-of-regulation}
\caption{同一棵句法树对应的不同规则推导}
\label{fig:2.5.2.4}
\end{figure}
......@@ -1138,7 +1128,7 @@ s_0 \overset{r_1}{\Rightarrow} s_1 \overset{r_2}{\Rightarrow} s_2 \overset{r_3}{
%图2.5.2.5
\begin{figure}[htp]
\centering
\includegraphics{./Chapter2/Figures/perspectives-of-expert-ordinary-and-syntactic-parser.pdf}
\input{./Chapter2/Figures/figure-perspectives-of-expert-ordinary-and-syntactic-parser}
\caption{如何选择最佳的句法分析结果 - 专家、普通人和句法分析器的视角}
\label{fig:2.5.2.5}
\end{figure}
......@@ -1149,7 +1139,7 @@ s_0 \overset{r_1}{\Rightarrow} s_1 \overset{r_2}{\Rightarrow} s_2 \overset{r_3}{
%图2.5.2.6
\begin{figure}[htp]
\centering
\includegraphics{./Chapter2/Figures/probability-values-corresponding-to-different-derivations.pdf}
\input{./Chapter2/Figures/figure-probability-values-corresponding-to-different-derivations}
\caption{不同推导（句法树）对应的概率值}
\label{fig:2.5.2.6}
\end{figure}
......@@ -1176,49 +1166,43 @@ s_0 \overset{r_1}{\Rightarrow} s_1 \overset{r_2}{\Rightarrow} s_2 \overset{r_3}{
\parinterval 概率上下文无关文法与传统上下文无关文法的区别在于，每条规则都会有一个概率，描述规则生成的可能性。具体来说，规则$\textrm{P}(\alpha \to \beta)$的概率可以被定义为：
\begin{equation}
\begin{eqnarray}
\textrm{P}(\alpha \to \beta)=\textrm{P}(\beta | \alpha)
\label{eq:2.5.3.1}
\end{equation}
\end{eqnarray}
即，在给定规则左部的情况下生成规则右部的可能性。进一步，在上下文无关文法中，因为上下文无关的属性，每条规则之间的使用都是相互独立的 \footnote[3]{如果是上下文有关文法，规则会形如 $\textrm{a}\alpha \textrm{b}\to \textrm{a}\beta \textrm{b}$，这时$\alpha \to \beta $的过程会依赖前后上下文\textrm{a}和\textrm{b}}。因此可以把\textrm{P(d)}分解为规则概率的乘积进行计算：
\begin{equation}
\begin{aligned}
\textrm{P}(d) & = \textrm{P}(r_1∘r_2∘…∘r_n) \\
& = \textrm{P}(r_1) \cdot \textrm{P}(r_2) \cdots \textrm{P}(r_n)
\begin{eqnarray}
\textrm{P}(d) & = & \textrm{P}(r_1 \cdot r_2 \cdot … \cdot r_n) \nonumber \\
& = & \textrm{P}(r_1) \cdot \textrm{P}(r_2) \cdots \textrm{P}(r_n)
\label{eq:2.5.3.1}
\end{aligned}
\end{equation}
\end{eqnarray}
\parinterval 这样我们就可以得到每个推导d的概率值。这个模型，可以很好的解释词串的生成过程。比如，对于规则集
\begin{equation} \nonumber
\begin{aligned}
r_3: & VV \to \text{吃} \\
r_4: & NN \to \text{鱼} \\
r_6: & VP \to VV NN
\end{aligned}
\end{equation}
\begin{eqnarray}
r_3: & VV \to \text{吃}\nonumber \\
r_4: & NN \to \text{鱼}\nonumber \\
r_6: & VP \to VV NN \nonumber
\end{eqnarray}
\parinterval 可以得到 $d_1=r_3 \cdot r_4 \cdot r_6$的概率为
\begin{equation}
\begin{aligned}
\textrm{P}(d_1) & = \textrm{P}(r_3) \cdot \textrm{P}(r_4) \cdot \textrm{P}(r_6) \\
& = \textrm{P}(VV \to \text{吃}) \cdot \textrm{P}(NN \to \text{鱼}) \cdots \textrm{P}(VP \to VV NN)
\begin{eqnarray}
\textrm{P}(d_1) & = &\textrm{P}(r_3) \cdot \textrm{P}(r_4) \cdot \textrm{P}(r_6)\nonumber \\
& = & \textrm{P}(VV \to \text{吃}) \cdot \textrm{P}(NN \to \text{鱼}) \cdots \textrm{P}(VP \to VV NN)
\label{eq:2.5.3.2}
\end{aligned}
\end{equation}
\end{eqnarray}
\parinterval 这也对应了词串“吃 鱼”的生成过程。首先，从起始VP使用规则$r_6$生成两个非终结符VV和NN；进一步，分别使用规则$r_3$和$r_4$从VV和NN进一步生成单词“吃”和“鱼”。整个过程的概率等于三条规则概率的乘积。
\parinterval 新的问题又来了，如何得到规则的概率呢？这里仍然可以使用数据驱动的想法，从数据中学习文法规则的概率。假设我们有人工标注的数据，其中包括很多句子的人工标注的句法树，称之为树库。然后，对于规则$\textrm{r}:\alpha \to \beta$可以使用极大似然估计：
\begin{equation}
\begin{eqnarray}
\textrm{P}(r) = \frac{\text{规则$r$在树库中出现的次数}}{\alpha \text{在树库中出现的次数}}
\label{eq:2.5.3.3}
\end{equation}
\end{eqnarray}
\parinterval 这里通过一个例子来解释规则概率的计算过程（图\ref{fig:2.5.3.1}）。
......@@ -1226,7 +1210,7 @@ r_6: & VP \to VV NN
% 图2.5.3.1
\begin{figure}[htp]
\centering
\includegraphics{./Chapter2/Figures/evaluation-of-probability-for-grammar.pdf}
\input{./Chapter2/Figures/figure-evaluation-of-probability-for-grammar}
\caption{上下文无关文法规则概率估计}
\label{fig:2.5.3.1}
\end{figure}
......@@ -1240,7 +1224,7 @@ r_6: & VP \to VV NN
% 图2.5.3.2
\begin{figure}[htp]
\centering
\includegraphics{./Chapter2/Figures/process-of-statistical-syntax-analysis.pdf}
\input{./Chapter2/Figures/figure-process-of-statistical-syntax-analysis}
\caption{统计句法分析的流程}
\label{fig:2.5.3.2}
\end{figure}
......@@ -1250,12 +1234,12 @@ r_6: & VP \to VV NN
%-------------------------------------------
% 图2.5.3.3
%\begin{figure}[htp]
% \centering
%\includegraphics{./Chapter2/Figures/.pdf}
% \caption{中文句法分析实例}
% \label{fig:2.5.3.3}
%\end{figure}
\begin{figure}[htp]
\centering
\input{./Chapter2/Figures/figure-example-of-zh-syntactic-analysis}
\caption{中文句法分析实例}
\label{fig:2.5.3.3}
\end{figure}
%-------------------------------------------
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
......
Book/Chapter3/Chapter3.tex
This source diff could not be displayed because it is too large. You can view the blob instead.
Book/Chapter3/Figures/figure-1.tex 0 → 100644
%%% outline
%-------------------------------------------------------------------------
\begin{tikzpicture}
\node [anchor=west,inner sep=2pt,fill=red!20,minimum height=3em] (eq1) at (0,0) {$f(s_u|t_v)$};
\node [anchor=west,inner sep=2pt] (eq2) at ([xshift=-2pt]eq1.east) {$=$};
\node [anchor=west,inner sep=2pt] (eq3) at ([xshift=-2pt]eq2.east) {$\lambda_{t_v}^{-1}$};
\node [anchor=west,inner sep=2pt] (eq4) at ([xshift=-2pt]eq3.east) {$\frac{\epsilon}{(l+1)^{m}}$};
\node [anchor=west,inner sep=2pt,fill=red!20,minimum height=3em] (eq5) at ([xshift=-2pt]eq4.east) {\footnotesize{$\prod\limits_{j=1}^{m} \sum\limits_{i=0}^{l} f(s_j|t_i)$}};
\node [anchor=west,inner sep=2pt] (eq6) at ([xshift=-2pt]eq5.east) {\footnotesize{$\sum\limits_{j=1}^{m} \delta(s_j,s_u) \sum\limits_{i=0}^{l} \delta(t_i,t_v)$}};
\node [anchor=west,inner sep=2pt,fill=red!20,minimum height=3em] (eq7) at ([xshift=-2pt,yshift=-0pt]eq6.east) {$\frac{f(s_u|t_v)}{\sum_{i=0}^{l}f(s_u|t_i)}$};
\node [anchor=south west,inner sep=2pt] (label1) at ([yshift=1em]eq1.north west) {\footnotesize{\textbf{新的参数值}}};
\node [anchor=south east,inner sep=2pt] (label2) at ([yshift=1em,xshift=-5em]eq7.north east) {\footnotesize{\textbf{旧的参数值}}};
\draw [<-,thick] (label1.south) .. controls +(south:1em) and +(north:1em) .. ([xshift=-1em]eq1.north);
\draw [<-,thick] (label2.south) .. controls +(300:1em) and +(north:1em) .. ([xshift=1em]eq7.north);
\draw [<-,thick] ([xshift=-0.5em]label2.south) .. controls +(240:1em) and +(north:1em) .. ([xshift=1em]eq5.north);
\end{tikzpicture}
%---------------------------------------------------------------------
Book/Chapter3/Figures/figure-2.tex 0 → 100644
%%% outline
%-------------------------------------------------------------------------
\begin{tikzpicture}
\node [anchor=west,inner sep=2pt,minimum height=2em] (eq1) at (0,0) {$f(s_u|t_v)$};
\node [anchor=west,inner sep=2pt] (eq2) at ([xshift=-2pt]eq1.east) {$=$};
\node [anchor=west,inner sep=2pt,minimum height=2em] (eq3) at ([xshift=-2pt]eq2.east) {$\lambda_{t_v}^{-1}$};
\node [anchor=west,inner sep=2pt,minimum height=3.0em] (eq4) at ([xshift=-3pt]eq3.east) {\footnotesize{$\frac{\epsilon}{(l+1)^{m}} \prod\limits_{j=1}^{m} \sum\limits_{i=0}^{l} f(s_j|t_i)$}};
\node [anchor=west,inner sep=2pt,minimum height=3.0em] (eq5) at ([xshift=1pt]eq4.east) {\footnotesize{$\sum\limits_{j=1}^{m} \delta(s_j,s_u) \sum\limits_{i=0}^{l} \delta(t_i,t_v)$}};
\node [anchor=west,inner sep=2pt,minimum height=3.0em] (eq6) at ([xshift=1pt]eq5.east) {$\frac{f(s_u|t_v)}{\sum_{i=0}^{l}f(s_u|t_i)}$};
{
\node [anchor=west,inner sep=2pt,fill=red!20,minimum height=3.0em] (eq4) at ([xshift=-3pt]eq3.east) {\footnotesize{$\frac{\epsilon}{(l+1)^{m}} \prod\limits_{j=1}^{m} \sum\limits_{i=0}^{l} f(s_j|t_i)$}};
}
{
\node [anchor=west,inner sep=2pt,fill=blue!20,minimum height=3.0em] (eq5) at ([xshift=1pt]eq4.east) {\footnotesize{$\sum\limits_{j=1}^{m} \delta(s_j,s_u) \sum\limits_{i=0}^{l} \delta(t_i,t_v)$}};
}
{
\node [anchor=west,inner sep=2pt,fill=green!20,minimum height=3.0em] (eq6) at ([xshift=1pt]eq5.east) {$\frac{f(s_u|t_v)}{\sum_{i=0}^{l}f(s_u|t_i)}$};
}
{
\node [anchor=south west,inner sep=2pt] (label1) at (eq4.north west) {\textbf{\scriptsize{翻译概率$\textrm{P}(s|t)$}}};
}
{
\node [anchor=south west,inner sep=2pt] (label2) at (eq5.north west) {\textbf{\scriptsize{配对的总次数}}};
\node [anchor=south west,inner sep=2pt] (label2part2) at ([yshift=-3pt]label2.north west) {\textbf{\scriptsize{$(s_u,t_v)$在句对$(s,t)$中}}};
}
{
\node [anchor=south west,inner sep=2pt] (label3) at (eq6.north west) {\textbf{\scriptsize{有的$t_i$的相对值}}};
\node [anchor=south west,inner sep=2pt] (label4) at ([yshift=-3pt]label3.north west) {\textbf{\scriptsize{$f(s_u|t_v)$对于所}}};
}
{
\node [anchor=east,rotate=90] (neweq1) at ([yshift=-0em]eq4.south) {=};
\node [anchor=north,inner sep=1pt] (neweq1full) at (neweq1.west) {\large{$\textrm{P}(s|t)$}};
}
{
\draw[decorate,thick,decoration={brace,amplitude=5pt,mirror}] ([yshift=-0.2em]eq5.south west) -- ([yshift=-0.2em]eq6.south east) node [pos=0.4,below,xshift=-0.0em,yshift=-0.3em] (expcount1) {\footnotesize{\textbf{'$t_v$翻译为$s_u$'这个事件}}};
\node [anchor=north west] (expcount2) at ([yshift=0.5em]expcount1.south west) {\footnotesize{\textbf{出现次数的期望的估计}}};
\node [anchor=north west] (expcount3) at ([yshift=0.5em]expcount2.south west) {\footnotesize{\textbf{称之为期望频次expected count}}};
}
\end{tikzpicture}
Book/Chapter3/Figures/figure-3.tex 0 → 100644
%%% outline
%-------------------------------------------------------------------------
\begin{tikzpicture}
\node [anchor=west,inner sep=2pt] (eq1) at (0,0) {$f(s_u|t_v)$};
\node [anchor=west] (eq2) at (eq1.east) {$=$\ };
\draw [-] ([xshift=0.3em]eq2.east) -- ([xshift=11.6em]eq2.east);
\node [anchor=south west] (eq3) at ([xshift=1em]eq2.east) {$\sum_{i=1}^{N} c_{\mathbb{E}}(s_u|t_v;s^{[i]},t^{[i]})$};
\node [anchor=north west] (eq4) at (eq2.east) {$\sum_{s_u} \sum_{i=1}^{N} c_{\mathbb{E}}(s_u|t_v;s^{[i]},t^{[i]})$};
{
\node [anchor=south] (label1) at ([yshift=-6em,xshift=3em]eq1.north west) {利用这个公式计算};
\node [anchor=north west] (label1part2) at ([yshift=0.3em]label1.south west) {新的$f(s_u|t_v)$值};
}
{
\node [anchor=west] (label2) at ([xshift=5em]label1.east) {用当前的$f(s_u|t_v)$};
\node [anchor=north west] (label2part2) at ([yshift=0.3em]label2.south west) {计算期望频次$c_{\mathbb{E}}(\cdot)$};
}
{
\node [anchor=west,fill=red!20,inner sep=2pt] (eq1) at (0,0) {$f(s_u|t_v)$};
}
\begin{pgfonlayer}{background}
{
\node[rectangle,fill=blue!20,inner sep=0] [fit = (eq3) (eq4)] (c) {};
}
{
\node[rectangle,draw,red,inner sep=0] [fit = (label1) (label1part2)] (flabel) {};
}
{
\node[rectangle,draw,ublue,inner sep=0] [fit = (label2) (label2part2)] (clabel) {};
}
\end{pgfonlayer}
{
\draw [->,thick] (eq1.south) ..controls +(south:1.5em) and +(north:1.5em).. (flabel.north);
}
{
\draw [->,thick] (c.south) ..controls +(south:1.0em) and +(north:1.0em).. (clabel.north);
}
{
\draw [->,thick] ([yshift=1em]flabel.east) -- ([yshift=1em]clabel.west);
\draw [<-,thick] ([yshift=-1em]flabel.east) -- ([yshift=-1em]clabel.west) node [pos=0.5,above,yshift=0.3em] {\footnotesize{\textbf{反复执行}}};
}
\end{tikzpicture}
Book/Chapter3/Figures/figure-4.tex 0 → 100644
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
%%% outline
%-------------------------------------------------------------------------
\begin{tikzpicture}
\node [anchor=north west] (line1) at (0,0) {\textbf{IBM模型1的训练（EM算法）}};
\node [anchor=north west] (line2) at ([yshift=-0.3em]line1.south west) {输入: 平行语料${(s^{[1]},t^{[1]}),...,(s^{[N]},t^{[N]})}$};
\node [anchor=north west] (line3) at ([yshift=-0.1em]line2.south west) {输出: 参数$f(\cdot|\cdot)$的最优值};
\node [anchor=north west] (line4) at ([yshift=-0.1em]line3.south west) {1: \textbf{Function} \textsc{TrainItWithEM}($\{(s^{[1]},t^{[1]}),...,(s^{[N]},t^{[N]})\}$) };
\node [anchor=north west] (line5) at ([yshift=-0.1em]line4.south west) {2: \ \ Initialize $f(\cdot|\cdot)$ \hspace{5em} $\rhd$ 比如给$f(\cdot|\cdot)$一个均匀分布};
\node [anchor=north west] (line6) at ([yshift=-0.1em]line5.south west) {3: \ \ Loop until $f(\cdot|\cdot)$ converges};
\node [anchor=north west] (line7) at ([yshift=-0.1em]line6.south west) {4: \ \ \ \ \textbf{foreach} $k = 1$ to $N$ \textbf{do}};
\node [anchor=north west] (line8) at ([yshift=-0.1em]line7.south west) {5: \ \ \ \ \ \ \ \footnotesize{$c_{\mathbb{E}}(s_u|t_v;s^{[k]},t^{[k]}) = \sum\limits_{j=1}^{|s^{[k]}|} \delta(s_j,s_u) \sum\limits_{i=0}^{|t^{[k]}|} \delta(t_i,t_v) \cdot \frac{f(s_u|t_v)}{\sum_{i=0}^{l}f(s_u|t_i)}$}\normalsize{}};
\node [anchor=north west] (line9) at ([yshift=-0.1em]line8.south west) {6: \ \ \ \ \textbf{foreach} $t_v$ appears at least one of $\{t^{[1]},...,t^{[N]}\}$ \textbf{do}};
\node [anchor=north west] (line10) at ([yshift=-0.1em]line9.south west) {7: \ \ \ \ \ \ \ $\lambda_{t_v}^{'} = \sum_{s_u} \sum_{k=1}^{N} c_{\mathbb{E}}(s_u|t_v;s^{[k]},t^{[k]})$};
\node [anchor=north west] (line11) at ([yshift=-0.1em]line10.south west) {8: \ \ \ \ \ \ \ \textbf{foreach} $s_u$ appears at least one of $\{s^{[1]},...,s^{[N]}\}$ \textbf{do}};
\node [anchor=north west] (line12) at ([yshift=-0.1em]line11.south west) {9: \ \ \ \ \ \ \ \ \ $f(s_u|t_v) = \sum_{k=1}^{N} c_{\mathbb{E}}(s_u|t_v;s^{[k]},t^{[k]}) \cdot (\lambda_{t_v}^{'})^{-1}$};
\node [anchor=north west] (line13) at ([yshift=-0.1em]line12.south west) {10: \ \textbf{return} $f(\cdot|\cdot)$};
\begin{pgfonlayer}{background}
{
\node[rectangle,draw=ublue, inner sep=0mm] [fit =(line1)(line2)(line3)(line4)(line5)(line6)(line7)(line11)(line8)(line9)(line13)] {};
}
\end{pgfonlayer}
\end{tikzpicture}
Book/Chapter3/Figures/figure325.tex → Book/Chapter3/Figures/figure-alignment-of-all-words-in-zh-en-sentence.tex
......@@ -4,146 +4,146 @@
%-------------------------------------------------------------------------
\begin{tikzpicture}
\begin{scope}
{\scriptsize
{
\node [anchor=west,inner sep=2pt] (s1) at (0,0) {谢谢};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=0.4em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.7em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.9em]s2.center) {you};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.4em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.6em]s2.center) {you};
\node [anchor=east,inner sep=2pt] (t0) at ([xshift=-0.2em,yshift=-0.05em]t1.west) {$t_0$};
\draw [-] (s1.south) -- (t0.north);
\draw [-] (s2.south) -- (t0.north);
{
\node [anchor=south east,inner sep=0pt] (p) at (t0.north west) {\small{{\color{ugreen} P(}}};
\node [anchor=south west,inner sep=0pt] (p2) at ([yshift=0.2em]t2.north east) {\small{{\color{ugreen} )}}};
\node [anchor=west] (eq) at (p2.east) {\small{+}};
\node [anchor=west] (eq) at ([xshift=0.7em]p2.east) {\small{+}};
}
}
\end{scope}
\begin{scope}[xshift=1.2in]
{\scriptsize
\begin{scope}[xshift=1.5in]
{
\node [anchor=west,inner sep=2pt] (s1) at (0,0) {谢谢};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=0.4em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.7em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.9em]s2.center) {you};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.4em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.6em]s2.center) {you};
\node [anchor=east,inner sep=2pt] (t0) at ([xshift=-0.2em,yshift=-0.05em]t1.west) {$t_0$};
\draw [-] (s1.south) -- (t0.north);
\draw [-] (s2.south) -- (t1.north);
{
\node [anchor=south east,inner sep=0pt] (p) at (t0.north west) {\small{{\color{ugreen} P(}}};
\node [anchor=south west,inner sep=0pt] (p2) at ([yshift=0.2em]t2.north east) {\small{{\color{ugreen} )}}};
\node [anchor=west] (eq) at (p2.east) {\small{+}};
\node [anchor=west] (eq) at ([xshift=0.7em]p2.east) {\small{+}};
}
}
\end{scope}
\begin{scope}[xshift=2.4in]
{\scriptsize
\begin{scope}[xshift=3in]
{
\node [anchor=west,inner sep=2pt] (s1) at (0,0) {谢谢};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=0.4em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.7em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.9em]s2.center) {you};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.4em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.6em]s2.center) {you};
\node [anchor=east,inner sep=2pt] (t0) at ([xshift=-0.2em,yshift=-0.05em]t1.west) {$t_0$};
\draw [-] (s1.south) -- (t0.north);
\draw [-] (s2.south) -- (t2.north);
{
\node [anchor=south east,inner sep=0pt] (p) at (t0.north west) {\small{{\color{ugreen} P(}}};
\node [anchor=south west,inner sep=0pt] (p2) at ([yshift=0.2em]t2.north east) {\small{{\color{ugreen} )}}};
\node [anchor=west] (eq) at (p2.east) {\small{+}};
\node [anchor=west] (eq) at ([xshift=0.7em]p2.east) {\small{+}};
}
}
\end{scope}
\begin{scope}[yshift=-0.5in]
{\scriptsize
\begin{scope}[yshift=-0.6in]
{
\node [anchor=west,inner sep=2pt] (s1) at (0,0) {谢谢};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=0.4em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.7em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.9em]s2.center) {you};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.4em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.6em]s2.center) {you};
\node [anchor=east,inner sep=2pt] (t0) at ([xshift=-0.2em,yshift=-0.05em]t1.west) {$t_0$};
\draw [-] (s1.south) -- ([yshift=-0.2em]t1.north);
\draw [-] (s2.south) -- (t0.north);
{
\node [anchor=south east,inner sep=0pt] (p) at (t0.north west) {\small{{\color{ugreen} P(}}};
\node [anchor=south west,inner sep=0pt] (p2) at ([yshift=0.2em]t2.north east) {\small{{\color{ugreen} )}}};
\node [anchor=west] (eq) at (p2.east) {\small{+}};
\node [anchor=west] (eq) at ([xshift=0.7em]p2.east) {\small{+}};
}
}
\end{scope}
\begin{scope}[xshift=1.2in,yshift=-0.5in]
{\scriptsize
\begin{scope}[xshift=1.5in,yshift=-0.6in]
{
\node [anchor=west,inner sep=2pt] (s1) at (0,0) {谢谢};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=0.4em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.7em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.9em]s2.center) {you};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.4em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.6em]s2.center) {you};
\node [anchor=east,inner sep=2pt] (t0) at ([xshift=-0.2em,yshift=-0.05em]t1.west) {$t_0$};
\draw [-] (s1.south) -- ([yshift=-0.2em]t1.north);
\draw [-] (s2.south) -- ([yshift=-0.2em]t1.north);
{
\node [anchor=south east,inner sep=0pt] (p) at (t0.north west) {\small{{\color{ugreen} P(}}};
\node [anchor=south west,inner sep=0pt] (p2) at ([yshift=0.2em]t2.north east) {\small{{\color{ugreen} )}}};
\node [anchor=west] (eq) at (p2.east) {\small{+}};
\node [anchor=west] (eq) at ([xshift=0.7em]p2.east) {\small{+}};
}
}
\end{scope}
\begin{scope}[xshift=2.4in,yshift=-0.5in]
{\scriptsize
\begin{scope}[xshift=3in,yshift=-0.6in]
{
\node [anchor=west,inner sep=2pt] (s1) at (0,0) {谢谢};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=0.4em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.7em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.9em]s2.center) {you};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.4em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.6em]s2.center) {you};
\node [anchor=east,inner sep=2pt] (t0) at ([xshift=-0.2em,yshift=-0.05em]t1.west) {$t_0$};
\draw [-] (s1.south) -- ([yshift=-0.2em]t1.north);
\draw [-] (s2.south) -- (t2.north);
{
\node [anchor=south east,inner sep=0pt] (p) at (t0.north west) {\small{{\color{ugreen} P(}}};
\node [anchor=south west,inner sep=0pt] (p2) at ([yshift=0.2em]t2.north east) {\small{{\color{ugreen} )}}};
\node [anchor=west] (eq) at (p2.east) {\small{+}};
\node [anchor=west] (eq) at ([xshift=0.7em]p2.east) {\small{+}};
}
}
\end{scope}
\begin{scope}[yshift=-1in]
{\scriptsize
\begin{scope}[yshift=-1.2in]
{
\node [anchor=west,inner sep=2pt] (s1) at (0,0) {谢谢};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=0.4em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.7em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.9em]s2.center) {you};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.4em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.6em]s2.center) {you};
\node [anchor=east,inner sep=2pt] (t0) at ([xshift=-0.2em,yshift=-0.05em]t1.west) {$t_0$};
\draw [-] (s1.south) -- (t2.north);
\draw [-] (s2.south) -- (t0.north);
{
\node [anchor=south east,inner sep=0pt] (p) at (t0.north west) {\small{{\color{ugreen} P(}}};
\node [anchor=south west,inner sep=0pt] (p2) at ([yshift=0.2em]t2.north east) {\small{{\color{ugreen} )}}};
\node [anchor=west] (eq) at (p2.east) {\small{+}};
\node [anchor=west] (eq) at ([xshift=0.7em]p2.east) {\small{+}};
}
}
\end{scope}
\begin{scope}[xshift=1.2in,yshift=-1in]
{\scriptsize
\begin{scope}[xshift=1.5in,yshift=-1.2in]
{
\node [anchor=west,inner sep=2pt] (s1) at (0,0) {谢谢};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=0.4em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.7em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.9em]s2.center) {you};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.4em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.6em]s2.center) {you};
\node [anchor=east,inner sep=2pt] (t0) at ([xshift=-0.2em,yshift=-0.05em]t1.west) {$t_0$};
\draw [-] (s1.south) -- (t2.north);
\draw [-] (s2.south) -- (t1.north);
{
\node [anchor=south east,inner sep=0pt] (p) at (t0.north west) {\small{{\color{ugreen} P(}}};
\node [anchor=south west,inner sep=0pt] (p2) at ([yshift=0.2em]t2.north east) {\small{{\color{ugreen} )}}};
\node [anchor=west] (eq) at (p2.east) {\small{+}};
\node [anchor=west] (eq) at ([xshift=0.7em]p2.east) {\small{+}};
}
}
\end{scope}
\begin{scope}[xshift=2.4in,yshift=-1in]
{\scriptsize
\begin{scope}[xshift=3in,yshift=-1.2in]
{
\node [anchor=west,inner sep=2pt] (s1) at (0,0) {谢谢};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=0.4em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.7em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.9em]s2.center) {you};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.4em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.6em]s2.center) {you};
\node [anchor=east,inner sep=2pt] (t0) at ([xshift=-0.2em,yshift=-0.05em]t1.west) {$t_0$};
\draw [-] (s1.south) -- (t2.north);
\draw [-] (s2.south) -- (t2.north);
{
\node [anchor=south east,inner sep=0pt] (p) at (t0.north west) {\small{{\color{ugreen} P(}}};
\node [anchor=south west,inner sep=0pt] (p2) at ([yshift=0.2em]t2.north east) {\small{{\color{ugreen} )}}};
\node [anchor=west] (eq) at (p2.east) {\normalsize{= \ P($s|t$)}};
\node [anchor=west] (eq) at ([xshift=0.7em]p2.east) {\normalsize{= \ P($s|t$)}};
}
}
\end{scope}
......
Book/Chapter3/Figures/figure322.tex → Book/Chapter3/Figures/figure-different-alignment-comparison.tex
......@@ -6,34 +6,34 @@
\begin{scope}
{\small
\node [anchor=west,inner sep=2pt] (s1) at (0,0) {谢谢};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=1em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.2em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.45em]s2.center) {you};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=2em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-2.2em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-2.45em]s2.center) {you};
\draw [-] (s1.south) -- ([yshift=-0.2em]t1.north);
\draw [-] (s2.south) -- (t2.north);
\node [anchor=center,draw=ublue,circle,thick,fill=white,inner sep=1pt,circular drop shadow={shadow xshift=0.1em,shadow yshift=-0.1em}] (mark) at ([xshift=0.5em,yshift=-0.2em]s2.south east) {{\color{ugreen} \tiny{\textbf{Yes}}}};
\node [anchor=center,draw=ublue,circle,thick,fill=white,inner sep=1pt,circular drop shadow={shadow xshift=0.1em,shadow yshift=-0.1em}] (mark) at ([xshift=0.8em,yshift=-0.7em]s2.south east) {{\color{ugreen} \tiny{\textbf{Yes}}}};
}
\end{scope}
\begin{scope}[xshift=1.3in]
\begin{scope}[xshift=1.8in]
{\small
\node [anchor=west,inner sep=2pt] (s1) at (0,0) {谢谢};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=1em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.2em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.45em]s2.center) {you};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=2em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-2.2em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-2.45em]s2.center) {you};
\draw [-] (s1.south) -- ([yshift=-0.2em]t1.north);
\draw [-] (s1.south) -- (t2.north);
\node [anchor=center,draw=ublue,circle,thick,fill=white,inner sep=1.5pt,circular drop shadow={shadow xshift=0.1em,shadow yshift=-0.1em}] (mark) at ([xshift=0.5em,yshift=-0.2em]s2.south east) {{\color{red} \tiny{\textbf{No}}}};
\node [anchor=center,draw=ublue,circle,thick,fill=white,inner sep=1.5pt,circular drop shadow={shadow xshift=0.1em,shadow yshift=-0.1em}] (mark) at ([xshift=0.8em,yshift=-0.7em]s2.south east) {{\color{red} \tiny{\textbf{No}}}};
}
\end{scope}
\begin{scope}[xshift=2.6in]
\begin{scope}[xshift=3.6in]
{\small
\node [anchor=west,inner sep=2pt] (s1) at (0,0) {谢谢};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=1em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-1.2em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-1.45em]s2.center) {you};
\node [anchor=west,inner sep=2pt] (s2) at ([xshift=2em]s1.east) {你};
\node [anchor=north,inner sep=2pt] (t1) at ([yshift=-2.2em]s1.center) {thank};
\node [anchor=north,inner sep=2pt] (t2) at ([yshift=-2.45em]s2.center) {you};
\draw [-] (s1.south) -- ([yshift=-0.2em]t1.north);
\draw [-] (s2.south) -- ([yshift=-0.2em]t1.north);
\node [anchor=center,draw=ublue,circle,thick,fill=white,inner sep=1pt,circular drop shadow={shadow xshift=0.1em,shadow yshift=-0.1em}] (mark) at ([xshift=0.5em,yshift=-0.2em]s2.south east) {{\color{ugreen} \tiny{\textbf{Yes}}}};
\node [anchor=center,draw=ublue,circle,thick,fill=white,inner sep=1pt,circular drop shadow={shadow xshift=0.1em,shadow yshift=-0.1em}] (mark) at ([xshift=0.8em,yshift=-0.7em]s2.south east) {{\color{ugreen} \tiny{\textbf{Yes}}}};
}
\end{scope}
\end{tikzpicture}
......
Book/Chapter3/Figures/figure319.tex → Book/Chapter3/Figures/figure-different-translation-candidate-space.tex
......@@ -13,7 +13,7 @@
\draw [->,thick,] (s1.north east) .. controls +(north east:1em) and +(north west:1em).. (t.north west) node[pos=0.5,below] {\tiny{正确翻译}};
\node [draw,red,fill=red!10,thick,anchor=center,circle,inner sep=3.5pt] (s) at ([xshift=10em,yshift=0em]s1.east) {\black{$s$}};
\node [draw,red,fill=red!10,thick,anchor=center,circle,inner sep=3.5pt] (s) at ([xshift=13em,yshift=0em]s1.east) {\black{$s$}};
\node [draw,ublue,fill=blue!10,thick,anchor=center,circle,inner sep=2pt] (t1) at ([xshift=1in]s.east) {\black{$t_1$}};
\node [draw,ublue,fill=blue!10,thick,anchor=center,circle,inner sep=2pt] (t2) at ([xshift=3em,yshift=2em]t1.north east) {\black{$t_2$}};
\node [draw,ublue,fill=blue!10,thick,anchor=center,circle,inner sep=2pt] (t3) at ([xshift=1em,yshift=4em]t1.north east) {\black{$t_3$}};
......@@ -28,6 +28,10 @@
\draw [->,thick,] (s.north) .. controls +(70:4em) and +(west:1em).. (t3.west) node[pos=0.5,above,xshift=-1em] {\tiny{P($t_3|s$)=0.3}};
\draw [->,thick,] (s.south east) .. controls +(300:3em) and +(south west:1em).. (t4.south west) node[pos=0.5,below] {\tiny{P($t_4|s$)=0.1}};
\node [anchor=center] (foot1) at ([xshift=3.8em,yshift=-3em]s1.south) {\footnotesize{人的翻译候选空间}};
\node [anchor=center] (foot2) at ([xshift=7em,yshift=-3em]s.south) {\footnotesize{机器的翻译候选空间}};
\end{tikzpicture}
......
Book/Chapter3/Figures/figure-example-of-formula1.29.tex 0 → 100644
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
%%% outline
%-------------------------------------------------------------------------
\begin{tikzpicture}
{\footnotesize
\node [anchor=west] (mid) at (0,0) {$\alpha(1,0)\alpha(2,0) + \alpha(1,0)\alpha(2,1) + \alpha(1,0)\alpha(2,2) +$};
\node [anchor=west] (mid2) at ([yshift=-2em]mid.west) {$\alpha(1,1)\alpha(2,0) + \alpha(1,1)\alpha(2,1) + \alpha(1,1)\alpha(2,2)+$};
\node [anchor=west] (mid3) at ([yshift=-2em]mid2.west) {$\alpha(1,2)\alpha(2,0) + \alpha(1,2)\alpha(2,1) + \alpha(1,2)\alpha(2,2)$};
}
\begin{pgfonlayer}{background}
\node[rectangle,draw=ublue,red,inner sep=0.1em,fill=white] [fit = (mid) (mid2) (mid3)] (exampleeq) {};
\end{pgfonlayer}
{\footnotesize
{
\node [anchor=north] (eq1) at ([xshift=2em,yshift=-2em]exampleeq.south west) {$\sum\limits_{y_1=0}^{2} \sum\limits_{y_2=0}^{2} \alpha(1,y_1)\alpha(2,y_2)$};
\node [anchor=west] (eq1part2) at ([xshift=-1em,yshift=-3em]eq1.west) {$=$};
\node [anchor=west] (eq1part3) at ([xshift=-0.5em]eq1part2.east) {$\sum\limits_{y_1=0}^{2} \sum\limits_{y_2=0}^{2} \prod\limits_{x=1}^{2} $};
\node [anchor=west,inner sep=2pt] (eq1part4) at ([xshift=-0.3em]eq1part3.east) {$\alpha(x,y_x)$};
}
{
\node [anchor=north] (eq2) at ([xshift=-2em,yshift=-2em]exampleeq.south east) {$(\alpha(1,0)+\alpha(1,1)+\alpha(1,2))\cdot$};
\node [anchor=west] (eq2part2) at ([yshift=-1.5em]eq2.west) {$(\alpha(2,0)+\alpha(2,1)+\alpha(2,2))$};
\node [anchor=west] (eq2part3) at ([xshift=2.1in]eq1part2.east){$=$};
\node [anchor=west] (eq2part4) at ([xshift=-0.5em]eq2part3.east){$\prod\limits_{x=1}^{2} \sum\limits_{y=0}^{2}$};
\node [anchor=west,inner sep=2pt] (eq2part5) at ([xshift=-0.3em]eq2part4.east){$\alpha(x,y)$};
}
}
\begin{pgfonlayer}{background}
{
\node[rectangle,draw=ublue,red,inner sep=0.1em,fill=white] [fit = (eq1) (eq1part2) (eq1part3)] (eq1full) {};
}
{
\node[rectangle,draw=ublue,red,inner sep=0.1em,fill=white] [fit = (eq2) (eq2part2) (eq2part3) (eq2part4)] (eq2full) {};
}
\end{pgfonlayer}
{
\draw [->,thick] ([xshift=-3em]exampleeq.south) .. controls +(south:1.5em) and +(north:1.5em) .. (eq1full.north);
}
{
\draw [->,thick] ([xshift=3em]exampleeq.south) .. controls +(south:1.5em) and +(north:1.5em) .. (eq2full.north);
}
{
\node [anchor=west] at ([xshift=0.7em]eq1full.east) {\LARGE{\textbf{=}}};
}
{
{\large
\node [anchor=west] (feq) at ([xshift=3em,yshift=-3em]eq1full.south west) {$\sum\limits_{a_1=0}^{l} ... \sum\limits_{a_m=0}^{l} \prod\limits_{j=1}^{m}$};
\node [anchor=west,inner sep=2pt,fill=blue!20] (feqpart2) at ([xshift=-0.3em]feq.east) {$f(s_j|t_{a_j})$};
\node [anchor=west,inner sep=1pt] (feqpart3) at (feqpart2.east) {=};
\node [anchor=west] (feqpart4) at (feqpart3.east) {$\prod\limits_{j=1}^{m} \sum\limits_{i=0}^{l}$};
\node [anchor=west,inner sep=2pt,fill=blue!20] (feqpart5) at ([xshift=-0.3em]feqpart4.east) {$f(s_j|t_i)$};
}
\draw [->,thick] (eq1part4.south) .. controls +(south:2.5em) and +(north:2.5em) .. (feqpart2.north);
\draw [->,thick] (eq2part5.south) .. controls +(south:1.5em) and +(north:1.5em) .. (feqpart5.north);
\node [anchor=west,inner sep=2pt,fill=blue!20] (eq1part4) at ([xshift=-0.3em]eq1part3.east) {\footnotesize{$\alpha(x,y_x)$}};
\node [anchor=west,inner sep=2pt,fill=blue!20] (eq2part5) at ([xshift=-0.3em]eq2part4.east){\footnotesize{$\alpha(x,y)$}};
}
\end{tikzpicture}
Book/Chapter3/Figures/figure312.tex → Book/Chapter3/Figures/figure-example-translation-alignment.tex
......@@ -14,21 +14,21 @@
{\footnotesize
\begin{scope}
\node [anchor=west] (s1) at (0,0) {我\footnotesize{$_1$}};
\node [anchor=west] (s2) at ([xshift=0.5em]s1.east) {对\footnotesize{$_2$}};
\node [anchor=west] (s3) at ([xshift=0.5em]s2.east) {你\footnotesize{$_3$}};
\node [anchor=west] (s4) at ([xshift=0.5em]s3.east) {感到\footnotesize{$_4$}};
\node [anchor=west] (s5) at ([xshift=0.5em]s4.east) {满意\footnotesize{$_5$}};
\node [anchor=west] (s1) at (0,0) {我$_1$};
\node [anchor=west] (s2) at ([xshift=2.2em]s1.east) {对$_2$};
\node [anchor=west] (s3) at ([xshift=3.2em]s2.east) {你$_3$};
\node [anchor=west] (s4) at ([xshift=3.6em]s3.east) {感到$_4$};
\node [anchor=west] (s5) at ([xshift=1.9em]s4.east) {满意$_5$};
\node [anchor=east] (s) at (s1.west) {$s=$};
\end{scope}
\begin{scope}[yshift=-2.6em]
\node [anchor=west] (t1) at (0.35em,0) {I\footnotesize{$_1$}};
\node [anchor=west] (t2) at ([xshift=0.3em,yshift=-0.1em]t1.east) {am\footnotesize{$_2$}};
\node [anchor=west] (t3) at ([xshift=0.3em,yshift=0.1em]t2.east) {satisfied\footnotesize{$_3$}};
\node [anchor=west] (t4) at ([xshift=0.3em]t3.east) {with\footnotesize{$_4$}};
\node [anchor=west] (t5) at ([xshift=0.3em,yshift=-0.2em]t4.east) {you\footnotesize{$_5$}};
\node [anchor=east] (t) at ([xshift=-0.3em]t1.west) {$t'=$};
\begin{scope}[yshift=-3.6em]
\node [anchor=west] (t1) at (0.35em,0) {I$_1$};
\node [anchor=west] (t2) at ([xshift=2.3em,yshift=-0.1em]t1.east) {am$_2$};
\node [anchor=west] (t3) at ([xshift=2.3em,yshift=0.1em]t2.east) {satisfied$_3$};
\node [anchor=west] (t4) at ([xshift=2.3em]t3.east) {with$_4$};
\node [anchor=west] (t5) at ([xshift=2.3em,yshift=-0.2em]t4.east) {you$_5$};
\node [anchor=east] (t) at (t1.west) {$t'=$};
\end{scope}
......@@ -51,20 +51,20 @@
{\footnotesize
\begin{scope}
\node [anchor=west] (s1) at (0,0) {我\footnotesize{$_1$}};
\node [anchor=west] (s2) at ([xshift=0.5em]s1.east) {对\footnotesize{$_2$}};
\node [anchor=west] (s3) at ([xshift=0.5em]s2.east) {你\footnotesize{$_3$}};
\node [anchor=west] (s4) at ([xshift=0.5em]s3.east) {感到\footnotesize{$_4$}};
\node [anchor=west] (s5) at ([xshift=0.5em]s4.east) {满意\footnotesize{$_5$}};
\node [anchor=west] (s1) at (0,0) {我$_1$};
\node [anchor=west] (s2) at ([xshift=2.5em]s1.east) {对$_2$};
\node [anchor=west] (s3) at ([xshift=2.5em]s2.east) {你$_3$};
\node [anchor=west] (s4) at ([xshift=2.5em]s3.east) {感到$_4$};
\node [anchor=west] (s5) at ([xshift=2.5em]s4.east) {满意$_5$};
\node [anchor=east] (s) at (s1.west) {$s=$};
\end{scope}
\begin{scope}[yshift=-2.6em]
\node [anchor=center] (t1) at ([yshift=-1.6em]s1.south) {I\footnotesize{$_1$}};
\node [anchor=center] (t2) at ([yshift=-1.6em]s2.south) {with\footnotesize{$_2$}};
\node [anchor=center] (t3) at ([yshift=-1.7em]s3.south) {you\footnotesize{$_3$}};
\node [anchor=center] (t4) at ([yshift=-1.7em]s4.south) {am\footnotesize{$_4$}};
\node [anchor=center] (t5) at ([yshift=-1.6em]s5.south) {satisfied\footnotesize{$_5$}};
\begin{scope}[yshift=-3.6em]
\node [anchor=center] (t1) at ([yshift=-1.6em]s1.south) {I$_1$};
\node [anchor=center] (t2) at ([yshift=-1.6em]s2.south) {with$_2$};
\node [anchor=center] (t3) at ([yshift=-1.7em]s3.south) {you$_3$};
\node [anchor=center] (t4) at ([yshift=-1.7em]s4.south) {am$_4$};
\node [anchor=center] (t5) at ([yshift=-1.6em]s5.south) {satisfied$_5$};
\node [anchor=center] (t) at ([xshift=-1.3em]t1.west) {$t''=$};
\end{scope}
......
Book/Chapter3/Figures/figure-greedy-MT-decoding-1.tex 0 → 100644
\begin{tikzpicture}
{
\node [anchor=north west,inner sep=2pt,align=left] (line1) at (0,0) {\textrm{\textbf{Function} \textsc{WordDecoding}($s$)}};
\node [anchor=north west,inner sep=2pt,align=left] (line2) at ([yshift=-1pt]line1.south west) {\textrm{1: $\pi = $\textsc{GetTransOptions}($s$)}};
\node [anchor=north west,inner sep=2pt,align=left] (line3) at ([yshift=-1pt]line2.south west) {\textrm{2: $best = \phi$}};
\node [anchor=north west,inner sep=2pt,align=left] (line4) at ([yshift=-1pt]line3.south west) {\textrm{3: \textbf{for} $i$ in $[1,m]$ \textbf{do}}};
\node [anchor=north west,inner sep=2pt,align=left] (line5) at ([yshift=-1pt]line4.south west) {\textrm{4: \hspace{1em} $h = \phi$}};
\node [anchor=north west,inner sep=2pt,align=left] (line6) at ([yshift=-1pt]line5.south west) {\textrm{5: \hspace{1em} \textbf{foreach} $j$ in $[1,m]$ \textbf{do}}};
\node [anchor=north west,inner sep=2pt,align=left] (line7) at ([yshift=-1pt]line6.south west) {\textrm{6: \hspace{2em} \textbf{if} $used[j]=$ \textbf{true} \textbf{then}}};
\node [anchor=north west,inner sep=2pt,align=left] (line8) at ([yshift=-1pt]line7.south west) {\textrm{7: \hspace{3em} $h = h \cup \textrm{\textsc{Join}}(best,\pi[j])$}};
\node [anchor=north west,inner sep=2pt,align=left] (line9) at ([yshift=-1pt]line8.south west) {\textrm{8: \hspace{1em} $best = \textrm{\textsc{PruneForTop1}}(h)$}};
\node [anchor=north west,inner sep=2pt,align=left] (line10) at ([yshift=-1pt]line9.south west) {\textrm{9: \hspace{1em} $used[best.j] = \textrm{\textsc{\textbf{true}}}$}};
\node [anchor=north west,inner sep=2pt,align=left] (line11) at ([yshift=-1pt]line10.south west) {\textrm{10: \textbf{return} $best.translatoin$}};
\node [anchor=south west,inner sep=2pt,align=left] (head1) at ([yshift=1pt]line1.north west) {输出: 找的最佳译文};
\node [anchor=south west,inner sep=2pt,align=left] (head2) at ([yshift=1pt]head1.north west) {输入: 源语句子$s=s_1...s_m$};
}
\begin{pgfonlayer}{background}
\node[rectangle,draw=ublue,thick,inner sep=0.2em,fill=white,drop shadow,minimum height=1.6cm] [fit = (head2) (line8) (line11)] (algorithm) {};
%% highlights
%\begin{pgfonlayer}{background}
{
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=2.21in] (line2highlight) at (line2.west) {};
}
{
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=2.21in] (line3highlight) at (line3.west) {};
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=2.21in] (line5highlight) at (line5.west) {};
}
%\end{pgfonlayer}
\end{pgfonlayer}
{
%% remark 1
\begin{scope}
{
\node [anchor=north west,align=left] (remark1) at ([xshift=0.4in]algorithm.north east) {获取每个单词\\的翻译候选};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,red] (s1) at ([yshift=-0.7em,xshift=0.5em]remark1.north east){1};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,ugreen] (s2) at ([xshift=0.4em]s1.east) {2};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,orange] (s3) at ([xshift=0.4em]s2.east) {3};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=3.0pt,ublue] (s4) at ([xshift=0.4em]s3.east) {...};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=1.5pt,purple] (s5) at ([xshift=0.4em]s4.east) {$m$};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,red,fill=red] (t1) at ([yshift=-1.7em]s1.center) {{\color{white} $n$}};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,ugreen,fill=ugreen] (t2) at ([yshift=-1.7em]s2.center) {{\color{white} $n$}};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,orange,fill=orange] (t3) at ([yshift=-1.7em]s3.center) {{\color{white} $n$}};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,ublue,fill=ublue] (t4) at ([yshift=-1.7em]s4.center) {{\color{white} $n$}};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,purple,fill=purple] (t5) at ([yshift=-1.7em]s5.center) {{\color{white} $n$}};
\draw [->,thick] ([yshift=-0.1em]s1.south) -- ([yshift=0.1em]t1.north);
\draw [->,thick] ([yshift=-0.1em]s2.south) -- ([yshift=0.1em]t2.north);
\draw [->,thick] ([yshift=-0.1em]s3.south) -- ([yshift=0.1em]t3.north);
\draw [->,thick] ([yshift=-0.1em]s4.south) -- ([yshift=0.1em]t4.north);
\draw [->,thick] ([yshift=-0.1em]s5.south) -- ([yshift=0.1em]t5.north);
\begin{pgfonlayer}{background}
{
\node[rectangle,draw,inner sep=0.2em,fill=blue!10] [fit = (remark1) (t5)] (remark1label) {};
}
\end{pgfonlayer}
}
\end{scope}
%% end of remark 1
%% remark 2
\begin{scope}
{
\node [anchor=north west,draw,inner sep=2pt,fill=blue!10] (remark2) at ([xshift=-0.2em,yshift=-1em]remark1.south west) {$best$用于保存当前最好的翻译结果};
}
\end{scope}
%% end of remark 2
\node [anchor=north west] (remark4) at ([xshift=21.8em,yshift=-0.6em]line7.east) {};
%% remark 3
\begin{scope}
{
\node [anchor=north west,draw,inner sep=2pt,fill=blue!10] (remark3) at ([yshift=-0.5em]remark2.south west) {$h$用于保存每步生成的所有译文候选};
}
\end{scope}
%% end of remark 3
{
\draw [->,thick] (line2highlight.east) ..controls +(east:1em) and +(west:1em).. (remark1label.west);
}
{
\draw [->,thick] (line3highlight.east) ..controls +(east:1em) and +(west:1em).. ([yshift=0.3em]remark2.south west);
\draw [->,thick] (line5highlight.east) ..controls +(east:1em) and +(west:1em).. ([yshift=0.3em]remark3.south west);
}
}
\end{tikzpicture}
Book/Chapter3/Figures/figure318.tex → Book/Chapter3/Figures/figure-greedy-MT-decoding-2.tex
......@@ -5,7 +5,7 @@
\begin{tikzpicture}
{\scriptsize
{
\node [anchor=north west,inner sep=2pt,align=left] (line1) at (0,0) {\textrm{\textbf{Function} \textsc{WordDecoding}($s$)}};
\node [anchor=north west,inner sep=2pt,align=left] (line2) at ([yshift=-1pt]line1.south west) {\textrm{1: $\pi = $\textsc{GetTransOptions}($s$)}};
......@@ -30,43 +30,43 @@
%% highlights
%\begin{pgfonlayer}{background}
{
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=1.71in] (line2highlight) at (line2.west) {};
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=2.21in] (line2highlight) at (line2.west) {};
}
{
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=1.71in] (line3highlight) at (line3.west) {};
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=1.71in] (line5highlight) at (line5.west) {};
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=2.21in] (line3highlight) at (line3.west) {};
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=2.21in] (line5highlight) at (line5.west) {};
}
{
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=1.71in] (line8highlight) at (line8.west) {};
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=2.21in] (line8highlight) at (line8.west) {};
}
{
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=1.71in] (line9highlight) at (line9.west) {};
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=2.21in] (line9highlight) at (line9.west) {};
}
{
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=1.71in] (line10highlight) at (line10.west) {};
\node[anchor=west,fill=blue!20,minimum height=0.16in,minimum width=2.21in] (line10highlight) at (line10.west) {};
}
%\end{pgfonlayer}
\end{pgfonlayer}
{\footnotesize
{
%% remark 1
\begin{scope}
{
\node [anchor=north west,align=left] (remark1) at ([xshift=0.4in]algorithm.north east) {获取每个单词\\的翻译候选};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,red] (s1) at ([yshift=-1.0em,xshift=0.5em]remark1.north east){\tiny{1}};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,ugreen] (s2) at ([xshift=0.4em]s1.east) {\tiny{2}};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,orange] (s3) at ([xshift=0.4em]s2.east) {\tiny{3}};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=3.0pt,ublue] (s4) at ([xshift=0.4em]s3.east) {.};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=1.5pt,purple] (s5) at ([xshift=0.4em]s4.east) {\tiny{$m$}};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,red] (s1) at ([yshift=-0.7em,xshift=0.5em]remark1.north east){1};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,ugreen] (s2) at ([xshift=0.4em]s1.east) {2};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,orange] (s3) at ([xshift=0.4em]s2.east) {3};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=3.0pt,ublue] (s4) at ([xshift=0.4em]s3.east) {...};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=1.5pt,purple] (s5) at ([xshift=0.4em]s4.east) {$m$};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,red,fill=red] (t1) at ([yshift=-1.7em]s1.center) {{\color{white} \tiny{$n$}}};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,ugreen,fill=ugreen] (t2) at ([yshift=-1.7em]s2.center) {{\color{white} \tiny{$n$}}};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,orange,fill=orange] (t3) at ([yshift=-1.7em]s3.center) {{\color{white} \tiny{$n$}}};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,ublue,fill=ublue] (t4) at ([yshift=-1.7em]s4.center) {{\color{white} \tiny{$n$}}};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,purple,fill=purple] (t5) at ([yshift=-1.7em]s5.center) {{\color{white} \tiny{$n$}}};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,red,fill=red] (t1) at ([yshift=-1.7em]s1.center) {{\color{white} $n$}};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,ugreen,fill=ugreen] (t2) at ([yshift=-1.7em]s2.center) {{\color{white} $n$}};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,orange,fill=orange] (t3) at ([yshift=-1.7em]s3.center) {{\color{white} $n$}};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,ublue,fill=ublue] (t4) at ([yshift=-1.7em]s4.center) {{\color{white} $n$}};
\node [anchor=center,draw,thick,circle,minimum size=0.3em,inner sep=2pt,purple,fill=purple] (t5) at ([yshift=-1.7em]s5.center) {{\color{white} $n$}};
\draw [->,thick] ([yshift=-0.1em]s1.south) -- ([yshift=0.1em]t1.north);
\draw [->,thick] ([yshift=-0.1em]s2.south) -- ([yshift=0.1em]t2.north);
......@@ -104,8 +104,8 @@
{
\node [anchor=north west,inner sep=2pt,align=left] (remark4) at ([xshift=0.25em,yshift=-0.6em]remark3.south west) {\textsc{Join}($a,b$) 返回\\$a$ 和$b$ 的所有组合};
{\scriptsize
\node [anchor=north west,inner sep=1pt,align=center,draw] (a1) at ([yshift=-0.5em]remark4.north east) {a1\\a2};
{
\node [anchor=north west,inner sep=1pt,align=center,draw] (a1) at ([yshift=-0.2em]remark4.north east) {a1\\a2};
\node [anchor=west] (join) at (a1.east) {$\times$};
\node [anchor=north west,inner sep=1pt,align=center,draw] (b1) at ([xshift=1.5em]a1.north east) {b1\\b2};
\node [anchor=west] (join) at (b1.east) {$=$};
......@@ -124,33 +124,33 @@
%% remark 5
\begin{scope}
{
\node [anchor=north west,align=left] (remark5) at ([xshift=0.0em,yshift=-1em]remark4.south west) {\textsc{PruneForTop1}\\保留得分最高的结果};
\node [anchor=west,draw,inner sep=1pt] (s1) at ([yshift=-0.5em,xshift=1.2em]remark5.north east){\tiny{0.234}};
\node [anchor=north west,draw,inner sep=1pt] (s2) at ([yshift=-0.2em]s1.south west){\tiny{0.197}};
\node [anchor=north west,draw,inner sep=1pt] (s3) at ([yshift=-0.2em]s2.south west){\tiny{0.083}};
\node [anchor=north west,align=left] (remark5) at ([xshift=0.0em,yshift=-1.3em]remark4.south west) {\textsc{PruneForTop1}\\保留得分最高的结果};
\node [anchor=west,draw,inner sep=1pt] (s1) at ([yshift=-0.5em,xshift=1.2em]remark5.north east){0.234};
\node [anchor=north west,draw,inner sep=1pt] (s2) at ([yshift=-0.2em]s1.south west){0.197};
\node [anchor=north west,draw,inner sep=1pt] (s3) at ([yshift=-0.2em]s2.south west){0.083};
\draw [-] ([yshift=-0.1em,xshift=-0.2em]s1.south west) -- ([yshift=-0.1em,xshift=3em]s1.south east);
\node [anchor=west] (top1) at ([xshift=0.1em]s1.east) {\tiny{\color{red}{$\gets$ top1}}};
\node [anchor=west] (top1) at ([xshift=0.1em]s1.east) {\color{red}{$\gets$ top1}};
\begin{pgfonlayer}{background}
{
\node[rectangle,draw,inner sep=0.2em,fill=blue!10] [fit = (remark5) (top1)] (remark5label) {};
\node[rectangle,draw,inner sep=0.2em,fill=blue!10] [fit = (remark5) (s3) (top1)] (remark5label) {};
}
\end{pgfonlayer}
}
\end{scope}
\%% end of remark 5
% end of remark 5
%% remark 6
\begin{scope}
{
\node [anchor=north west,align=left] (remark6) at ([xshift=0.0em,yshift=-1em]remark5.south west) {记录已经翻译过\\的源语单词};
\node [anchor=north west,align=left] (remark6) at ([xshift=0.0em,yshift=-1.3em]remark5.south west) {记录已经翻译过\\的源语单词};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,red] (s1) at ([yshift=-1.0em,xshift=0.5em]remark6.north east){\tiny{1}};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,ugreen] (s2) at ([xshift=0.4em]s1.east) {\tiny{2}};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,orange] (s3) at ([xshift=0.4em]s2.east) {\tiny{3}};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=3.0pt,ublue] (s4) at ([xshift=0.4em]s3.east) {.};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=1.5pt,purple] (s5) at ([xshift=0.4em]s4.east) {\tiny{$m$}};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,red] (s1) at ([yshift=-1.3em,xshift=0.5em]remark6.north east){1};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,ugreen] (s2) at ([xshift=0.4em]s1.east) {2};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=2.1pt,orange] (s3) at ([xshift=0.4em]s2.east) {3};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=3.0pt,ublue] (s4) at ([xshift=0.4em]s3.east) {...};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,inner sep=1.5pt,purple] (s5) at ([xshift=0.4em]s4.east) {$m$};
\draw [-,thick,red] (s1.north east) -- (s1.south west);
\draw [-,thick,orange] (s3.north east) -- (s3.south west);
......@@ -162,7 +162,7 @@
\end{pgfonlayer}
}
\end{scope}
\%% end of remark 6
% end of remark 6
{
\draw [->,thick] (line2highlight.east) ..controls +(east:1em) and +(west:1em).. (remark1label.west);
......
Book/Chapter3/Figures/figure33.tex → Book/Chapter3/Figures/figure-human-translation.tex
......@@ -86,11 +86,11 @@
\node [anchor=north west] (label1) at (ft11.south west) {\small{选择最佳单词翻译，调整词序，得到完美的结果}};
}
{
\draw[decorate,thick,decoration={brace,amplitude=5pt,mirror}] ([yshift=8em,xshift=-0.5em]t13.south west) -- ([xshift=-0.5em]t13.south west) node [pos=0.5,left,xshift=-0.5em,yshift=0.5em] (label2) {\footnotesize{\textbf{学习到的}}};
\draw[decorate,thick,decoration={brace,amplitude=5pt,mirror}] ([yshift=8em,xshift=-0.7em]t13.south west) -- ([xshift=-0.7em]t13.south west) node [pos=0.5,left,xshift=-0.5em,yshift=0.5em] (label2) {\footnotesize{\textbf{学习到的}}};
\node [anchor=north west] (label2part2) at ([yshift=0.3em]label2.south west) {\footnotesize{\textbf{单词翻译}}};
}
{
\draw[decorate,thick,decoration={brace,amplitude=5pt,mirror}] ([yshift=-0.2em,xshift=-0.5em]t13.south west) -- ([yshift=-5em,xshift=-0.5em]t13.south west) node [pos=0.5,left,xshift=-0.5em,yshift=0.5em] (label3) {\footnotesize{\textbf{运用知识}}};
\draw[decorate,thick,decoration={brace,amplitude=5pt,mirror}] ([yshift=-0.2em,xshift=-0.7em]t13.south west) -- ([yshift=-5em,xshift=-0.7em]t13.south west) node [pos=0.5,left,xshift=-0.5em,yshift=0.5em] (label3) {\footnotesize{\textbf{运用知识}}};
\node [anchor=north west] (label3part2) at ([yshift=0.3em]label3.south west) {\footnotesize{\textbf{生成译文}}};
}
\end{scope}
......
Book/Chapter3/Figures/figure35.tex → Book/Chapter3/Figures/figure-process-of-machine-translation.tex
%%% outline
%-------------------------------------------------------------------------
\begin{tikzpicture}
\begin{scope}
......@@ -160,21 +157,45 @@
\end{scope}
\begin{scope}
{\footnotesize
\node [anchor=east] (label4) at ([yshift=0.8em]ft11.west) {翻译就是一条};
\node [anchor=north west] (label4part2) at ([yshift=0.7em]label4.south west) {译文选择路径};
}
{\footnotesize
\node [anchor=east] (label5) at ([yshift=0.4em]ft21.west) {不同的译文对};
\node [anchor=north west] (label5part2) at ([yshift=0.7em]label5.south west) {应不同的路径};
}
{\footnotesize
\node [anchor=east] (label6) at ([yshift=0.4em]ft31.west) {单词翻译的词};
\node [anchor=north west] (label6part2) at ([yshift=0.7em]label6.south west) {序也可能不同};
}
{\footnotesize
\node [anchor=east] (label7) at ([yshift=0.4em]ft41.west) {可能的翻译路};
\node [anchor=north west] (label7part2) at ([yshift=0.7em]label7.south west) {径非常多};
}
\end{scope}
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
\begin{scope}
{
\draw[decorate,thick,decoration={brace,amplitude=5pt,mirror}] ([yshift=8em,xshift=-0.5em]t13.south west) -- ([xshift=-0.5em]t13.south west) node [pos=0.5,left,xshift=-0.5em,yshift=2.0em] (label2) {\footnotesize{\textbf{从双语数}}};
\draw[decorate,thick,decoration={brace,amplitude=5pt}] ([yshift=8em,xshift=2.0em]t53.south east) -- ([xshift=2.0em]t53.south east) node [pos=0.5,right,xshift=0.5em,yshift=2.0em] (label2) {\footnotesize{\textbf{从双语数}}};
\node [anchor=north west] (label2part2) at ([yshift=0.3em]label2.south west) {\footnotesize{\textbf{据中自动}}};
\node [anchor=north west] (label2part3) at ([yshift=0.3em]label2part2.south west) {\footnotesize{\textbf{学习词典}}};
\node [anchor=north west] (label2part4) at ([yshift=0.3em]label2part3.south west) {\footnotesize{\textbf{(训练)}}};
}
{
\draw[decorate,thick,decoration={brace,amplitude=5pt,mirror}] ([yshift=-0.2em,xshift=-0.5em]t13.south west) -- ([yshift=-9.5em,xshift=-0.5em]t13.south west) node [pos=0.5,left,xshift=-0.5em,yshift=2.0em] (label3) {\footnotesize{\textbf{利用概率}}};
\draw[decorate,thick,decoration={brace,amplitude=5pt}] ([yshift=-1.0em,xshift=6.2em]t53.south west) -- ([yshift=-10.5em,xshift=6.2em]t53.south west) node [pos=0.5,right,xshift=0.5em,yshift=2.0em] (label3) {\footnotesize{\textbf{利用概率}}};
\node [anchor=north west] (label3part2) at ([yshift=0.3em]label3.south west) {\footnotesize{\textbf{化的词典}}};
\node [anchor=north west] (label3part3) at ([yshift=0.3em]label3part2.south west) {\footnotesize{\textbf{进行翻译}}};
\node [anchor=north west] (label3part4) at ([yshift=0.3em]label3part3.south west) {\footnotesize{\textbf{(解码)}}};
}
\end{scope}
\begin{scope}
{
......@@ -195,14 +216,11 @@
\begin{scope}
{
\draw [->,ultra thick,ublue,line width=2pt,opacity=0.7] ([xshift=-0.5em]t13.west) -- ([xshift=0.8em]t13.east) -- ([xshift=-0.2em]t21.west) -- ([xshift=0.8em]t21.east) -- ([xshift=-0.2em]t31.west) -- ([xshift=0.8em]t31.east) -- ([xshift=-0.2em]t41.west) -- ([xshift=0.8em]t41.east) -- ([xshift=-0.2em]t51.west) -- ([xshift=1.2em]t51.east);
\draw [->,ultra thick,ublue,line width=2pt,opacity=0.7] ([xshift=-0.5em,yshift=-0.3em]t13.west) -- ([xshift=0.8em,yshift=-0.3em]t13.east) -- ([xshift=-0.2em,yshift=-0.3em]t21.west) -- ([xshift=0.8em,yshift=-0.3em]t21.east) -- ([xshift=-0.2em,yshift=-0.3em]t31.west) -- ([xshift=0.8em,yshift=-0.3em]t31.east) -- ([xshift=-0.2em,yshift=-0.3em]t41.west) -- ([xshift=0.8em,yshift=-0.3em]t41.east) -- ([xshift=-0.2em,yshift=-0.3em]t51.west) -- ([xshift=1.2em,yshift=-0.3em]t51.east);
}
\draw [->,ultra thick,red,line width=2pt,opacity=0.7] ([xshift=-0.5em]t13.west) -- ([xshift=0.8em]t13.east) -- ([xshift=-0.2em]t22.west) -- ([xshift=0.8em]t22.east) -- ([xshift=-0.2em]t31.west) -- ([xshift=0.8em]t31.east) -- ([xshift=-0.2em]t41.west) -- ([xshift=0.8em]t41.east) -- ([xshift=-0.2em]t52.west) -- ([xshift=1.2em]t52.east);
\draw [->,ultra thick,red,line width=2pt,opacity=0.7] ([xshift=-0.5em,yshift=-0.5em]t13.west) -- ([xshift=0.8em,yshift=-0.5em]t13.east) -- ([xshift=-0.2em,yshift=-0.5em]t22.west) -- ([xshift=0.8em,yshift=-0.5em]t22.east) -- ([xshift=-0.2em,yshift=-0.5em]t31.west) -- ([xshift=0.8em,yshift=-0.5em]t31.east) -- ([xshift=-0.2em,yshift=-0.5em]t41.west) -- ([xshift=0.8em,yshift=-0.5em]t41.east) -- ([xshift=-0.2em,yshift=-0.5em]t52.west) -- ([xshift=1.2em,yshift=-0.5em]t52.east);
\end{scope}
\end{tikzpicture}
\ No newline at end of file
%---------------------------------------------------------------------
Book/Chapter3/Figures/figure36.tex → Book/Chapter3/Figures/figure-processes-SMT.tex
......@@ -5,7 +5,7 @@
\begin{tikzpicture}
{\scriptsize
\node [anchor=north west,inner sep=1pt] (entry1) at (0,0) {\tiny{\textbf{1:} 这 是 数据 $\leftrightarrow$ This is data}};
\node [anchor=north west,inner sep=1pt] (entry1) at (0,0) {\tiny{\textbf{1:} 这 是 数据 $\leftrightarrow$ This is data \; \;}};
\node [anchor=north west,inner sep=1pt] (entry2) at ([yshift=0.1em]entry1.south west) {\tiny{\textbf{2:} 小心 ！$\leftrightarrow$ Look out !}};
\node [anchor=north west,inner sep=1pt] (entry3) at ([yshift=0.1em]entry2.south west) {\tiny{\textbf{3:} 你 是 谁 $\leftrightarrow$ Who are you}};
\node [anchor=north west,inner sep=2pt] (entry4) at ([yshift=0.1em]entry3.south west) {...};
......@@ -29,9 +29,9 @@
\draw [->,very thick,ublue] ([xshift=0.4em]model.east) -- ([xshift=3.4em]model.east) node [inner sep=0pt,pos=0.5, above,yshift=0.3em] (decodingarrow) {\color{red}{\scriptsize{穷举\&计算}}};
{\scriptsize
\node [anchor=north west,inner sep=2pt] (sentlabel) at ([xshift=5.5em,yshift=-0.3em]model.north east) {{\color{ublue} \textbf{机器翻译引擎}}};
\node [anchor=north west] (sent) at ([yshift=-0.5em]sentlabel.south west) {\textbf{对任意句子}};
\node [anchor=north west] (sentpart2) at ([yshift=0.3em]sent.south west) {\textbf{进行翻译}};
\node [anchor=north west,inner sep=2pt] (sentlabel) at ([xshift=5.5em,yshift=-0.9em]model.north east) {{\color{ublue} \textbf{机器翻译引擎}}};
\node [anchor=north west] (sent) at ([yshift=-0.5em]sentlabel.south west) {\textbf{对任意句子进行翻译}};
\node [anchor=north west] (sentpart2) at ([yshift=0.3em]sent.south west) {\textbf{\quad}};
}
}
......@@ -43,3 +43,22 @@
\end{tikzpicture}
%---------------------------------------------------------------------
Book/Chapter3/Figures/figure314.tex → Book/Chapter3/Figures/figure-scores-of-different-translation_model&language_model.tex
......@@ -10,21 +10,21 @@
{\footnotesize
\begin{scope}
\node [anchor=west] (s1) at (0,0) {我\footnotesize{$_1$}};
\node [anchor=west] (s2) at ([xshift=0.5em]s1.east) {对\footnotesize{$_2$}};
\node [anchor=west] (s3) at ([xshift=0.5em]s2.east) {你\footnotesize{$_3$}};
\node [anchor=west] (s4) at ([xshift=0.5em]s3.east) {感到\footnotesize{$_4$}};
\node [anchor=west] (s5) at ([xshift=0.5em]s4.east) {满意\footnotesize{$_5$}};
\node [anchor=west] (s1) at (0,0) {我$_1$};
\node [anchor=west] (s2) at ([xshift=2.2em]s1.east) {对$_2$};
\node [anchor=west] (s3) at ([xshift=3.2em]s2.east) {你$_3$};
\node [anchor=west] (s4) at ([xshift=3.6em]s3.east) {感到$_4$};
\node [anchor=west] (s5) at ([xshift=1.9em]s4.east) {满意$_5$};
\node [anchor=east] (s) at (s1.west) {$s=$};
\end{scope}
\begin{scope}[yshift=-2.6em]
\node [anchor=west] (t1) at (0.35em,0) {I\footnotesize{$_1$}};
\node [anchor=west] (t2) at ([xshift=0.3em,yshift=-0.1em]t1.east) {am\footnotesize{$_2$}};
\node [anchor=west] (t3) at ([xshift=0.3em,yshift=0.1em]t2.east) {satisfied\footnotesize{$_3$}};
\node [anchor=west] (t4) at ([xshift=0.3em]t3.east) {with\footnotesize{$_4$}};
\node [anchor=west] (t5) at ([xshift=0.3em,yshift=-0.2em]t4.east) {you\footnotesize{$_5$}};
\node [anchor=east] (t) at ([xshift=-0.3em]t1.west) {$t'=$};
\begin{scope}[yshift=-3.6em]
\node [anchor=west] (t1) at (0.35em,0) {I$_1$};
\node [anchor=west] (t2) at ([xshift=2.3em,yshift=-0.1em]t1.east) {am$_2$};
\node [anchor=west] (t3) at ([xshift=2.3em,yshift=0.1em]t2.east) {satisfied$_3$};
\node [anchor=west] (t4) at ([xshift=2.3em]t3.east) {with$_4$};
\node [anchor=west] (t5) at ([xshift=2.3em,yshift=-0.2em]t4.east) {you$_5$};
\node [anchor=east] (t) at (t1.west) {$t'=$};
\end{scope}
......@@ -47,20 +47,20 @@
{\footnotesize
\begin{scope}
\node [anchor=west] (s1) at (0,0) {我\footnotesize{$_1$}};
\node [anchor=west] (s2) at ([xshift=0.5em]s1.east) {对\footnotesize{$_2$}};
\node [anchor=west] (s3) at ([xshift=0.5em]s2.east) {你\footnotesize{$_3$}};
\node [anchor=west] (s4) at ([xshift=0.5em]s3.east) {感到\footnotesize{$_4$}};
\node [anchor=west] (s5) at ([xshift=0.5em]s4.east) {满意\footnotesize{$_5$}};
\node [anchor=west] (s1) at (0,0) {我$_1$};
\node [anchor=west] (s2) at ([xshift=2.5em]s1.east) {对$_2$};
\node [anchor=west] (s3) at ([xshift=2.5em]s2.east) {你$_3$};
\node [anchor=west] (s4) at ([xshift=2.5em]s3.east) {感到$_4$};
\node [anchor=west] (s5) at ([xshift=2.5em]s4.east) {满意$_5$};
\node [anchor=east] (s) at (s1.west) {$s=$};
\end{scope}
\begin{scope}[yshift=-2.6em]
\node [anchor=center] (t1) at ([yshift=-1.6em]s1.south) {I\footnotesize{$_1$}};
\node [anchor=center] (t2) at ([yshift=-1.6em]s2.south) {with\footnotesize{$_2$}};
\node [anchor=center] (t3) at ([yshift=-1.7em]s3.south) {you\footnotesize{$_3$}};
\node [anchor=center] (t4) at ([yshift=-1.7em]s4.south) {am\footnotesize{$_4$}};
\node [anchor=center] (t5) at ([yshift=-1.6em]s5.south) {satisfied\footnotesize{$_5$}};
\begin{scope}[yshift=-3.6em]
\node [anchor=center] (t1) at ([yshift=-1.6em]s1.south) {I$_1$};
\node [anchor=center] (t2) at ([yshift=-1.6em]s2.south) {with$_2$};
\node [anchor=center] (t3) at ([yshift=-1.7em]s3.south) {you$_3$};
\node [anchor=center] (t4) at ([yshift=-1.7em]s4.south) {am$_4$};
\node [anchor=center] (t5) at ([yshift=-1.6em]s5.south) {satisfied$_5$};
\node [anchor=center] (t) at ([xshift=-1.3em]t1.west) {$t''=$};
\end{scope}
......
Book/Chapter3/Figures/figure37.tex → Book/Chapter3/Figures/figure-simple-statistical-algorithms.tex
......@@ -4,16 +4,16 @@
\begin{tikzpicture}
\begin{scope}
{
\node [pos=0.5,left,xshift=-0.5em,yshift=2.0em] (label2) {\footnotesize{\textbf{Input: a word pair (x, y) and a sentence pair (s, t)}}};
\node [anchor=north west] (label2part2) at ([yshift=0.3em]label2.south west) {\footnotesize{\textbf{Output: the number of (x, y) in the (s, t)}}};
\node [anchor=north west] (label2part3) at ([yshift=0.3em]label2part2.south west) {\footnotesize{{1：count $\leftarrow$ 0}}};
\node [anchor=north west] (label2part4) at ([yshift=0.3em]label2part3.south west) {\footnotesize{{2：\textbf{for} s\underline{\hbox to 0.1cm{}}word $\leftarrow$ $S_1$ … $S_length(s)$ \textbf{do} }}};
\node [anchor=north west] (label2part5)at ([yshift=0.3em]label2part4.south west) {\footnotesize{{3：\quad\textbf{for}t\underline{\hbox to 0.1cm{}}word $\leftarrow$ $t_1$ … $t_length(t)$ \textbf{do} }}};
\node [anchor=north west] (label2part6)at ([yshift=0.3em]label2part5.south west) {\footnotesize{{4：\quad\quad\textbf{if} s\underline{\hbox to 0.1cm{}}word == x \textbf{and} t\underline{\hbox to 0.1cm{}}word==y \textbf{then} }}};
\node [anchor=north west] (label2part7) at ([yshift=0.3em]label2part6.south west) {\footnotesize{{5：\quad\quad\quad\quad count $\leftarrow$ count + 1 }}};
\node [anchor=north west] (label2part8) at ([yshift=0.3em]label2part7.south west) {\footnotesize{{6：\quad\quad \textbf{end if} }}};
\node [anchor=north west] (label2part9) at ([yshift=0.3em]label2part8.south west) {\footnotesize{{7：\quad \textbf{end for} }}};
\node [anchor=north west] (label2part10) at ([yshift=0.3em]label2part9.south west) {\footnotesize{{8： \textbf{end for} }}};
\node [pos=0.5,left,xshift=-0.5em,yshift=2.0em] (label2) {Input: a word pair (x, y) and a sentence pair (s, t)};
\node [anchor=north west] (label2part2) at ([yshift=0.3em]label2.south west) {Output: the number of (x, y) in the (s, t)};
\node [anchor=north west] (label2part3) at ([yshift=0.3em]label2part2.south west) {{1：$count$ $\leftarrow$ 0}};
\node [anchor=north west] (label2part4) at ([yshift=0.3em]label2part3.south west) {{2：\textbf{for} $s\underline{\hbox to 0.1cm{}}word$ $\leftarrow$ $S_1$ … $S_{length(s)}$ \textbf{do} }};
\node [anchor=north west] (label2part5)at ([yshift=0.3em]label2part4.south west) {{3：\quad\textbf{for} $t\underline{\hbox to 0.1cm{}}word$ $\leftarrow$ $t_1$ … $t_{length(t)}$ \textbf{do} }};
\node [anchor=north west] (label2part6)at ([yshift=0.3em]label2part5.south west) {{4：\quad\quad\textbf{if} $s\underline{\hbox to 0.1cm{}}word$ == x \textbf{and} $t\underline{\hbox to 0.1cm{}}word$==y \textbf{then} }};
\node [anchor=north west] (label2part7) at ([yshift=0.3em]label2part6.south west) {{5：\quad\quad\quad\quad $count$ $\leftarrow$ $count$ + 1 }};
\node [anchor=north west] (label2part8) at ([yshift=0.3em]label2part7.south west) {{6：\quad\quad \textbf{end if} }};
\node [anchor=north west] (label2part9) at ([yshift=0.3em]label2part8.south west) {{7：\quad \textbf{end for} }};
\node [anchor=north west] (label2part10) at ([yshift=0.3em]label2part9.south west) {{8： \textbf{end for} }};
}
......
Book/Chapter3/Figures/figure32.tex → Book/Chapter3/Figures/figure-translation-pipeline.tex
......@@ -4,31 +4,32 @@
\definecolor{ugreen}{rgb}{0,0.5,0}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\node [anchor=west,draw,thick,minimum width=7.7em,minimum height=0.8em] (sent) at (0,0) {};
\node [anchor=west] (sent) at (0,0) {\scriptsize{源文：我对你感到满意}};
\node [anchor=west,draw,thick,minimum width=10.5em,minimum height=1.2em] (sent-1) at (0,0) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,red] (s1) at ([yshift=-1.5em]sent.south west) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,ugreen] (s2) at ([xshift=0.7em]s1.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,orange] (s3) at ([xshift=0.7em]s2.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,ublue] (s4) at ([xshift=0.7em]s3.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,purple] (s5) at ([xshift=0.7em]s4.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,red] (s1) at ([xshift=0.15em,yshift=-2.7em]sent.south west) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,ugreen] (s2) at ([xshift=1.3em]s1.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,orange] (s3) at ([xshift=1.3em]s2.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,ublue] (s4) at ([xshift=1.3em]s3.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,purple] (s5) at ([xshift=1.3em]s4.east) {};
{
\node [anchor=west,draw,thick,circle,minimum size=0.3em,red,fill=red] (t1) at ([yshift=-2.0em]s1.west) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,ugreen,fill=ugreen] (t2) at ([xshift=0.7em]t1.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,orange,fill=orange] (t3) at ([xshift=0.7em]t2.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,ublue,fill=ublue] (t4) at ([xshift=0.7em]t3.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,purple,fill=purple] (t5) at ([xshift=0.7em]t4.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,red,fill=red] (t1) at ([yshift=-2.5em]s1.west) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,ugreen,fill=ugreen] (t2) at ([xshift=1.3em]t1.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,orange,fill=orange] (t3) at ([xshift=1.3em]t2.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,ublue,fill=ublue] (t4) at ([xshift=1.3em]t3.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,purple,fill=purple] (t5) at ([xshift=1.3em]t4.east) {};
}
{
\node [anchor=west,draw,thick,circle,minimum size=0.3em,red,fill=red] (ft1) at ([yshift=-2.0em]t1.west) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,ublue,fill=ublue] (ft2) at ([xshift=0.7em]ft1.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,purple,fill=purple] (ft3) at ([xshift=0.7em]ft2.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,ugreen,fill=ugreen] (ft4) at ([xshift=0.7em]ft3.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,orange,fill=orange] (ft5) at ([xshift=0.7em]ft4.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,red,fill=red] (ft1) at ([yshift=-2.5em]t1.west) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,ublue,fill=ublue] (ft2) at ([xshift=1.3em]ft1.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,purple,fill=purple] (ft3) at ([xshift=1.3em]ft2.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,ugreen,fill=ugreen] (ft4) at ([xshift=1.3em]ft3.east) {};
\node [anchor=west,draw,thick,circle,minimum size=0.3em,orange,fill=orange] (ft5) at ([xshift=1.3em]ft4.east) {};
}
\draw [->,thick,double] ([yshift=-0.1em]sent.south) -- ([yshift=-0.8em]sent.south);
\draw [->,thick,double] ([yshift=-0.1em]sent-1.south) -- ([yshift=-1.0em]sent-1.south);
{
\draw [->,thick] ([yshift=-0.1em]s1.south) -- ([yshift=0.1em]t1.north);
......@@ -45,13 +46,68 @@
\draw [->,thick] ([yshift=-0.1em]t5.south) -- ([yshift=0.1em]ft3.north);
}
{
\node [anchor=north west] (label1) at ([xshift=0.5em,yshift=0.3em]sent.south east) {{\scriptsize \textbf{分析}}};
\node [anchor=north west] (label2) at ([yshift=-0.5em]label1.south west) {{\scriptsize \textbf{转换}}};
\node [anchor=north west] (label3) at ([yshift=-0.5em]label2.south west) {{\scriptsize \textbf{生成}}};
\node [anchor=north west] (label1) at ([xshift=0.6em,yshift=0.0em]sent-1.south east) {{\scriptsize \textbf{分析}}};
\node [anchor=north west] (label2) at ([yshift=-1.8em]label1.south west) {{\scriptsize \textbf{转换}}};
\node [anchor=north west] (label3) at ([yshift=-1.3em]label2.south west) {{\scriptsize \textbf{生成}}};
}
{\scriptsize
\begin{scope}
\node [anchor=west] (ss1) at ([xshift=-1.6em,yshift=1.5em]s1.east) {我};
\node [anchor=west] (ss2) at ([xshift=1.4em]ss1.east) {对};
\node [anchor=west] (ss3) at ([xshift=1.5em]ss2.east) {你};
\node [anchor=west] (ss4) at ([xshift=1.0em]ss3.east) {感到};
\node [anchor=west] (ss5) at ([xshift=0.4em]ss4.east) {满意};
\end{scope}
\begin{scope}[yshift=-3em]
\node [anchor=west] (tt1) at ([xshift=-1.2em,yshift=-1.5em]ft1.east) {I};
\node [anchor=west] (tt2) at ([xshift=1.7em,yshift=-0.1em]tt1.east) {am};
\node [anchor=west] (tt3) at ([xshift=0.3em,yshift=0.1em]tt2.east) {satisfied};
\node [anchor=west] (tt4) at ([xshift=-0.1em]tt3.east) {with};
\node [anchor=west] (tt5) at ([xshift=0.7em,yshift=-0.2em]tt4.east) {you};
\end{scope}
}
\node [anchor=west] (sent2) at ([xshift=-3.0em,yshift=-1.8em]tt2.south) {\scriptsize{译文：I am satisfied with you }};
\node [anchor=west,draw,thick,minimum width=10.5em,minimum height=1.2em] (sent-2) at ([xshift=-3.0em,yshift=-1.8em]tt2.south) {};
\draw [->,thick,double] ([yshift=1.0em]sent-2.north) -- ([yshift=0.1em]sent-2.north);
\begin{pgfonlayer}{background}
{
%\node[rectangle,draw=ublue,thick, inner sep=0mm] [fit =(sent)] {};
%\node[rectangle,draw=ublue,thick, inner sep=0mm] [fit =(sent2)] {};
}
\end{pgfonlayer}
\end{tikzpicture}
%---------------------------------------------------------------------
Book/Chapter3/Figures/figure31.tex → Book/Chapter3/Figures/figure-zh-en-translation-example.tex
\begin{center}
%%% outline
%-------------------------------------------------------------------------
{\footnotesize
{
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\begin{tikzpicture}
......@@ -9,8 +9,8 @@
\node [anchor=west] (s1) at (0,0) {我};
\node [anchor=west] (s2) at ([xshift=0.8em]s1.east) {对};
\node [anchor=west] (s3) at ([xshift=0.8em]s2.east) {你};
\node [anchor=west] (s4) at ([xshift=0.8em]s3.east) {感到};
\node [anchor=west] (s5) at ([xshift=0.8em]s4.east) {满意};
\node [anchor=west] (s4) at ([xshift=0.9em]s3.east) {感到};
\node [anchor=west] (s5) at ([xshift=1.2em]s4.east) {满意};
\end{scope}
\begin{scope}[yshift=-3em]
......@@ -32,4 +32,3 @@
}
%---------------------------------------------------------------------
\end{center}
Book/Chapter3/Figures/figure000.tex 0 → 100644
\def\CTeXPreproc{Created by ctex v0.2.13, don't edit!}
\documentclass[cjk,t,compress,12pt]{standalone}
%\documentclass{article}
%\usepackage{beamerarticle}
\usepackage{pstricks}
\usepackage{etex}
\usepackage{eso-pic,graphicx}
\usepackage{fancybox}
\usepackage{amsmath,amssymb}
\usepackage{setspace}
\usepackage{xcolor}
\usepackage{CJK}
\usepackage{tikz}
\usepackage{tikz-qtree}
\usepackage{hyperref}
\usetikzlibrary{arrows,decorations.pathreplacing}
\usetikzlibrary{shadows} % LATEX and plain TEX when using Tik Z
\usepgflibrary{arrows} % LATEX and plain TEX and pure pgf
\usetikzlibrary{arrows} % LATEX and plain TEX when using Tik Z
\usetikzlibrary{decorations}
\usetikzlibrary{arrows,shapes}
\usetikzlibrary{decorations.text}
\usetikzlibrary{positioning,fit,calc}
\usetikzlibrary{mindmap,backgrounds} % mind map
\DeclareMathOperator*{\argmax}{arg\,max}
\DeclareMathOperator*{\argmin}{arg\,min}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\definecolor{ugreen}{rgb}{0,0.5,0}
\begin{document}
\begin{CJK}{UTF8}{you}
%%% outline
%-------------------------------------------------------------------------
\begin{tikzpicture}
{\footnotesize
\node [anchor=west] (mid) at (0,0) {$\alpha(1,0)\alpha(2,0) + \alpha(1,0)\alpha(2,1) + \alpha(1,0)\alpha(2,2) +$};
\node [anchor=west] (mid2) at ([yshift=-2em]mid.west) {$\alpha(1,1)\alpha(2,0) + \alpha(1,1)\alpha(2,1) + \alpha(1,1)\alpha(2,2)+$};
\node [anchor=west] (mid3) at ([yshift=-2em]mid2.west) {$\alpha(1,2)\alpha(2,0) + \alpha(1,2)\alpha(2,1) + \alpha(1,2)\alpha(2,2)$};
}
\begin{pgfonlayer}{background}
\node[rectangle,draw=ublue,red,inner sep=0.1em,fill=white] [fit = (mid) (mid2) (mid3)] (exampleeq) {};
\end{pgfonlayer}
{\footnotesize
{
\node [anchor=north] (eq1) at ([xshift=2em,yshift=-2em]exampleeq.south west) {$\sum\limits_{y_1=0}^{2} \sum\limits_{y_2=0}^{2} \alpha(1,y_1)\alpha(2,y_2)$};
\node [anchor=west] (eq1part2) at ([xshift=-1em,yshift=-3em]eq1.west) {$=$};
\node [anchor=west] (eq1part3) at ([xshift=-0.5em]eq1part2.east) {$\sum\limits_{y_1=0}^{2} \sum\limits_{y_2=0}^{2} \prod\limits_{x=1}^{2} $};
\node [anchor=west,inner sep=2pt] (eq1part4) at ([xshift=-0.3em]eq1part3.east) {$\alpha(x,y_x)$};
}
{
\node [anchor=north] (eq2) at ([xshift=-2em,yshift=-2em]exampleeq.south east) {$(\alpha(1,0)+\alpha(1,1)+\alpha(1,2))\cdot$};
\node [anchor=west] (eq2part2) at ([yshift=-1.5em]eq2.west) {$(\alpha(2,0)+\alpha(2,1)+\alpha(2,2))$};
\node [anchor=west] (eq2part3) at ([xshift=2.1in]eq1part2.east){$=$};
\node [anchor=west] (eq2part4) at ([xshift=-0.5em]eq2part3.east){$\prod\limits_{x=1}^{2} \sum\limits_{y=0}^{2}$};
\node [anchor=west,inner sep=2pt] (eq2part5) at ([xshift=-0.3em]eq2part4.east){$\alpha(x,y)$};
}
}
\begin{pgfonlayer}{background}
{
\node[rectangle,draw=ublue,red,inner sep=0.1em,fill=white] [fit = (eq1) (eq1part2) (eq1part3)] (eq1full) {};
}
{
\node[rectangle,draw=ublue,red,inner sep=0.1em,fill=white] [fit = (eq2) (eq2part2) (eq2part3) (eq2part4)] (eq2full) {};
}
\end{pgfonlayer}
{
\draw [->,thick] ([xshift=-3em]exampleeq.south) .. controls +(south:1.5em) and +(north:1.5em) .. (eq1full.north);
}
{
\draw [->,thick] ([xshift=3em]exampleeq.south) .. controls +(south:1.5em) and +(north:1.5em) .. (eq2full.north);
}
{
\node [anchor=west] at ([xshift=0.7em]eq1full.east) {\LARGE{\textbf{=}}};
}
{
{\large
\node [anchor=west] (feq) at ([xshift=3em,yshift=-3em]eq1full.south west) {$\sum\limits_{a_1=0}^{l} ... \sum\limits_{a_m=0}^{l} \prod\limits_{j=1}^{m}$};
\node [anchor=west,inner sep=2pt,fill=blue!20] (feqpart2) at ([xshift=-0.3em]feq.east) {$f(s_j|t_{a_j})$};
\node [anchor=west,inner sep=1pt] (feqpart3) at (feqpart2.east) {=};
\node [anchor=west] (feqpart4) at (feqpart3.east) {$\prod\limits_{j=1}^{m} \sum\limits_{i=0}^{l}$};
\node [anchor=west,inner sep=2pt,fill=blue!20] (feqpart5) at ([xshift=-0.3em]feqpart4.east) {$f(s_j|t_i)$};
}
\draw [->,thick] (eq1part4.south) .. controls +(south:2.5em) and +(north:2.5em) .. (feqpart2.north);
\draw [->,thick] (eq2part5.south) .. controls +(south:1.5em) and +(north:1.5em) .. (feqpart5.north);
\node [anchor=west,inner sep=2pt,fill=blue!20] (eq1part4) at ([xshift=-0.3em]eq1part3.east) {\footnotesize{$\alpha(x,y_x)$}};
\node [anchor=west,inner sep=2pt,fill=blue!20] (eq2part5) at ([xshift=-0.3em]eq2part4.east){\footnotesize{$\alpha(x,y)$}};
}
\end{tikzpicture}
%---------------------------------------------------------------------
\end{CJK}
\end{document}
Book/Chapter3/Figures/figure315.tex
......@@ -5,14 +5,14 @@
\definecolor{ugreen}{rgb}{0,0.5,0}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s1) at (0,0) {$\phi$};
\node [anchor=west,inner sep=2pt,fill=red!20,thick,minimum width=3.6em,minimum height=0.8em] (s1) at (0,0) {$\phi$};
{%第一列
\node [anchor=north west] (label1) at ([xshift=2.5em,yshift=6.3em]s1.east) {{\scriptsize \textbf{第1步}}};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s21) at ([xshift=2.0em,yshift=3.5em]s1.east) {$w_1^1$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s22) at ([xshift=2.0em,yshift=0.0em]s1.east) {$w_2^1$};
\node [anchor=west,inner sep=2pt,fill=green!20,thick,minimum width=3.6em,minimum height=0.8em] (s21) at ([xshift=2.0em,yshift=3.5em]s1.east) {$w_1^1$};
\node [anchor=west,inner sep=2pt,fill=green!20,thick,minimum width=3.6em,minimum height=0.8em] (s22) at ([xshift=2.0em,yshift=0.0em]s1.east) {$w_2^1$};
\node [anchor=west] (s23) at ([xshift=3.3em,yshift=-3.2em]s1.east) {$\vdots$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s24) at ([xshift=2.0em,yshift=-7.0em]s1.east) {\scriptsize{$w_{m \bullet n}^1$}};
\node [anchor=west,inner sep=2pt,fill=green!20,thick,minimum width=3.6em,minimum height=0.8em] (s24) at ([xshift=2.0em,yshift=-7.0em]s1.east) {\scriptsize{$w_{m \bullet n}^1$}};
}
{
\draw [->,thick] ([yshift=0.0em]s1.east) -- ([yshift=0.0em]s21.west);
......@@ -23,10 +23,10 @@
{%第二列
\node [anchor=north west] (label2) at ([xshift=9.3em,yshift=6.3em]s1.east) {{\scriptsize \textbf{第2步}}};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s31) at ([xshift=8.8em,yshift=3.5em]s1.east) {$w_1^2$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s32) at ([xshift=8.8em,yshift=0.0em]s1.east) {$w_2^2$};
\node [anchor=west,inner sep=2pt,fill=blue!20,thick,minimum width=3.6em,minimum height=0.8em] (s31) at ([xshift=8.8em,yshift=3.5em]s1.east) {$w_1^2$};
\node [anchor=west,inner sep=2pt,fill=blue!20,thick,minimum width=3.6em,minimum height=0.8em] (s32) at ([xshift=8.8em,yshift=0.0em]s1.east) {$w_2^2$};
\node [anchor=west] (s33) at ([xshift=10.1em,yshift=-3.2em]s1.east) {$\vdots$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s34) at ([xshift=8.8em,yshift=-7.0em]s1.east) {\tiny{$w_{(m-1) \bullet n}^2$}};
\node [anchor=west,inner sep=2pt,fill=blue!20,thick,minimum width=3.6em,minimum height=0.8em] (s34) at ([xshift=8.8em,yshift=-7.0em]s1.east) {\tiny{$w_{(m-1) \bullet n}^2$}};
}
{
\draw [->,thick] ([yshift=0.0em]s21.east) -- ([yshift=0.0em]s31.west);
......@@ -79,10 +79,10 @@
{%第四列
\node [anchor=north west] (label3) at ([xshift=19.4em,yshift=6.3em]s1.east) {{\scriptsize \textbf{第m步}}};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s51) at ([xshift=19.0em,yshift=3.5em]s1.east) {$w_1^m$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s52) at ([xshift=19.0em,yshift=0.0em]s1.east) {$w_2^m$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s53) at ([xshift=19.0em,yshift=-3.5em]s1.east) {$w_{?}^{m}$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s54) at ([xshift=19.0em,yshift=-7.0em]s1.east) {\scriptsize{$w_{1 \bullet n}^m$}};
\node [anchor=west,inner sep=2pt,fill=orange!20,thick,minimum width=3.6em,minimum height=0.8em] (s51) at ([xshift=19.0em,yshift=3.5em]s1.east) {$w_1^m$};
\node [anchor=west,inner sep=2pt,fill=orange!20,thick,minimum width=3.6em,minimum height=0.8em] (s52) at ([xshift=19.0em,yshift=0.0em]s1.east) {$w_2^m$};
\node [anchor=west,inner sep=2pt,fill=orange!20,thick,minimum width=3.6em,minimum height=0.8em] (s53) at ([xshift=19.0em,yshift=-3.5em]s1.east) {$w_{?}^{m}$};
\node [anchor=west,inner sep=2pt,fill=orange!20,thick,minimum width=3.6em,minimum height=0.8em] (s54) at ([xshift=19.0em,yshift=-7.0em]s1.east) {\scriptsize{$w_{1 \bullet n}^m$}};
\draw [->,thick] ([yshift=0.0em]s41.east) -- ([yshift=0.0em]s51.west);
\draw [->,thick] ([yshift=0.0em]s42.east) -- ([yshift=0.0em]s52.west);
......
Book/Chapter3/Figures/figure316.tex deleted 100644 → 0
%%% outline
%-------------------------------------------------------------------------
\begin{tabular}{l | l | l}
句子长度$m$ & 单词翻译候选数量$n$ & 译文数量$n^m \cdot m!$ \\ \hline
1 & 1 & 1 \\
1 & 10 & 10 \\
2 & 10 & 200 \\
10 & 10 & 36288000000000000 \\
20 & 10 & 2.43290200817664 $\times 10^{38}$ \\
20 & 30 & 8.48300477127188 $\times 10^{47}$ \\
\end{tabular}
%---------------------------------------------------------------------
Book/Chapter3/Figures/figure317.tex
......@@ -6,14 +6,14 @@
\definecolor{ugreen}{rgb}{0,0.5,0}
\definecolor{ublue}{rgb}{0.152,0.250,0.545}
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s1) at (0,0) {$\phi$};
\node [anchor=west,inner sep=2pt,fill=red!20,thick,minimum width=3.6em,minimum height=0.8em] (s1) at (0,0) {$\phi$};
{%第一列
\node [anchor=north west] (label1) at ([xshift=2.5em,yshift=6.3em]s1.east) {{\scriptsize \textbf{第1步}}};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s21) at ([xshift=2.0em,yshift=3.5em]s1.east) {$w_1^1$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s22) at ([xshift=2.0em,yshift=0.0em]s1.east) {$w_2^1$};
\node [anchor=west,inner sep=2pt,fill=green!20,thick,minimum width=3.6em,minimum height=0.8em] (s21) at ([xshift=2.0em,yshift=3.5em]s1.east) {$w_1^1$};
\node [anchor=west,inner sep=2pt,fill=green!20,thick,minimum width=3.6em,minimum height=0.8em] (s22) at ([xshift=2.0em,yshift=0.0em]s1.east) {$w_2^1$};
\node [anchor=west] (s23) at ([xshift=3.3em,yshift=-3.2em]s1.east) {$\vdots$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s24) at ([xshift=2.0em,yshift=-7.0em]s1.east) {\scriptsize{$w_{m \bullet n}^1$}};
\node [anchor=west,inner sep=2pt,fill=green!20,thick,minimum width=3.6em,minimum height=0.8em] (s24) at ([xshift=2.0em,yshift=-7.0em]s1.east) {\scriptsize{$w_{m \bullet n}^1$}};
}
{
\draw [->,thick,red] ([yshift=0.0em]s1.east) -- ([yshift=0.0em]s21.west);
......@@ -24,10 +24,10 @@
{%第二列
\node [anchor=north west] (label2) at ([xshift=9.3em,yshift=6.3em]s1.east) {{\scriptsize \textbf{第2步}}};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s31) at ([xshift=8.8em,yshift=3.5em]s1.east) {$w_1^2$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s32) at ([xshift=8.8em,yshift=0.0em]s1.east) {$w_2^2$};
\node [anchor=west,inner sep=2pt,fill=blue!20,thick,minimum width=3.6em,minimum height=0.8em] (s31) at ([xshift=8.8em,yshift=3.5em]s1.east) {$w_1^2$};
\node [anchor=west,inner sep=2pt,fill=blue!20,thick,minimum width=3.6em,minimum height=0.8em] (s32) at ([xshift=8.8em,yshift=0.0em]s1.east) {$w_2^2$};
\node [anchor=west] (s33) at ([xshift=10.1em,yshift=-3.2em]s1.east) {$\vdots$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s34) at ([xshift=8.8em,yshift=-7.0em]s1.east) {\tiny{$w_{(m-1) \bullet n}^2$}};
\node [anchor=west,inner sep=2pt,fill=blue!20,thick,minimum width=3.6em,minimum height=0.8em] (s34) at ([xshift=8.8em,yshift=-7.0em]s1.east) {\tiny{$w_{(m-1) \bullet n}^2$}};
}
{
\draw [->,thick,densely dashed] ([yshift=0.0em]s21.east) -- ([yshift=0.0em]s31.west);
......@@ -51,10 +51,10 @@
{%第四列
\node [anchor=north west] (label3) at ([xshift=19.4em,yshift=6.3em]s1.east) {{\scriptsize \textbf{第m步}}};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s51) at ([xshift=19.0em,yshift=3.5em]s1.east) {$w_1^m$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s52) at ([xshift=19.0em,yshift=0.0em]s1.east) {$w_2^m$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s53) at ([xshift=19.0em,yshift=-3.5em]s1.east) {$w_{?}^{m}$};
\node [anchor=west,draw,thick,minimum width=3.6em,minimum height=0.8em] (s54) at ([xshift=19.0em,yshift=-7.0em]s1.east) {\scriptsize{$w_{1 \bullet n}^m$}};
\node [anchor=west,inner sep=2pt,fill=orange!20,thick,minimum width=3.6em,minimum height=0.8em] (s51) at ([xshift=19.0em,yshift=3.5em]s1.east) {$w_1^m$};
\node [anchor=west,inner sep=2pt,fill=orange!20,thick,minimum width=3.6em,minimum height=0.8em] (s52) at ([xshift=19.0em,yshift=0.0em]s1.east) {$w_2^m$};
\node [anchor=west,inner sep=2pt,fill=orange!20,thick,minimum width=3.6em,minimum height=0.8em] (s53) at ([xshift=19.0em,yshift=-3.5em]s1.east) {$w_{?}^{m}$};
\node [anchor=west,inner sep=2pt,fill=orange!20,thick,minimum width=3.6em,minimum height=0.8em] (s54) at ([xshift=19.0em,yshift=-7.0em]s1.east) {\scriptsize{$w_{1 \bullet n}^m$}};
\draw [->,thick,densely dashed] ([yshift=0.0em]s41.east) -- ([yshift=0.0em]s51.west);
\draw [->,thick,densely dashed] ([yshift=0.0em]s42.east) -- ([yshift=0.0em]s52.west);
......
Book/Chapter3/Figures/figure310.tex → Book/Chapter3/Figures/figurerole-of-P-in-sentence-level-translation.tex
......@@ -5,7 +5,7 @@
\begin{tikzpicture}
{\scriptsize
\node [anchor=north west,inner sep=1pt] (entry1) at (0,0) {\tiny{\textbf{1:} 这 是 数据 $\leftrightarrow$ This is data}};
\node [anchor=north west,inner sep=1pt] (entry1) at (0,0) {\tiny{\textbf{1:} 这 是 数据 $\leftrightarrow$ This is data \; \;}};
\node [anchor=north west,inner sep=1pt] (entry2) at ([yshift=0.1em]entry1.south west) {\tiny{\textbf{2:} 小心 ！$\leftrightarrow$ Look out !}};
\node [anchor=north west,inner sep=1pt] (entry3) at ([yshift=0.1em]entry2.south west) {\tiny{\textbf{3:} 你 是 谁 $\leftrightarrow$ Who are you}};
\node [anchor=north west,inner sep=2pt] (entry4) at ([yshift=0.1em]entry3.south west) {...};
......@@ -29,8 +29,8 @@
{\scriptsize
\node [anchor=north west,inner sep=2pt] (sentlabel) at ([xshift=5.5em,yshift=-0.3em]model.north east) {{\color{ublue} \textbf{机器翻译引擎}}};
\node [anchor=north west] (sent) at ([yshift=-0.5em]sentlabel.south west) {\textbf{对任意句子}};
\node [anchor=north west] (sentpart2) at ([yshift=0.3em]sent.south west) {\textbf{进行翻译}};
\node [anchor=north west] (sent) at ([yshift=-0.5em]sentlabel.south west) {\textbf{对任意句子进行翻译}};
\node [anchor=north west] (sentpart2) at ([yshift=0.3em]sent.south west) {\textbf{\quad}};
}
\begin{pgfonlayer}{background}
......
Book/bibliography.bib
......@@ -298,3 +298,8 @@ year={2015}
year={1972},
publisher={Prentice-Hall Englewood Cliffs, NJ}
}
@inproceedings{huang2008advanced,
title={Advanced Dynamic Programming in CL},
author={Huang, Liang},
year={2008}
}
\ No newline at end of file
Book/mt-book.bbl 0 → 100644
% $ biblatex auxiliary file $
% $ biblatex bbl format version 3.1 $
% Do not modify the above lines!
%
% This is an auxiliary file used by the 'biblatex' package.
% This file may safely be deleted. It will be recreated by
% biber as required.
%
\begingroup
\makeatletter
\@ifundefined{ver@biblatex.sty}
{\@latex@error
{Missing 'biblatex' package}
{The bibliography requires the 'biblatex' package.}
\aftergroup\endinput}
{}
\endgroup
\refsection{0}
\datalist[entry]{nyt/global//global/global}
\entry{aho1972theory}{book}{}
\name{author}{2}{}{%
{{hash=87e9142329f09390b038692270dc2168}{%
family={Aho},
familyi={A\bibinitperiod},
given={Alfred\bibnamedelima V},
giveni={A\bibinitperiod\bibinitdelim V\bibinitperiod}}}%
{{hash=cafe1d770c56825ebccdb81e9f7c9b3b}{%
family={Ullman},
familyi={U\bibinitperiod},
given={Jeffrey\bibnamedelima D},
giveni={J\bibinitperiod\bibinitdelim D\bibinitperiod}}}%
}
\list{publisher}{1}{%
{Prentice-Hall Englewood Cliffs, NJ}%
}
\strng{namehash}{012da699a378a549fed2220d3c78ed22}
\strng{fullhash}{012da699a378a549fed2220d3c78ed22}
\strng{bibnamehash}{012da699a378a549fed2220d3c78ed22}
\strng{authorbibnamehash}{012da699a378a549fed2220d3c78ed22}
\strng{authornamehash}{012da699a378a549fed2220d3c78ed22}
\strng{authorfullhash}{012da699a378a549fed2220d3c78ed22}
\field{sortinit}{A}
\field{sortinithash}{a3dcedd53b04d1adfd5ac303ecd5e6fa}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{The theory of parsing, translation, and compiling}
\field{volume}{1}
\field{year}{1972}
\endentry
\entry{bahdanau2014neural}{article}{}
\name{author}{3}{}{%
{{hash=6d80adec79a13a33e73215c5f46f1605}{%
family={Bahdanau},
familyi={B\bibinitperiod},
given={Dzmitry},
giveni={D\bibinitperiod}}}%
{{hash=3da7501a79d9346572c7fd6e41b615df}{%
family={Cho},
familyi={C\bibinitperiod},
given={Kyunghyun},
giveni={K\bibinitperiod}}}%
{{hash=40a8e4774982146adc2688546f54efb2}{%
family={Bengio},
familyi={B\bibinitperiod},
given={Yoshua},
giveni={Y\bibinitperiod}}}%
}
\strng{namehash}{ccf5ebef61998aaab5ec6eace8f4564d}
\strng{fullhash}{ccf5ebef61998aaab5ec6eace8f4564d}
\strng{bibnamehash}{ccf5ebef61998aaab5ec6eace8f4564d}
\strng{authorbibnamehash}{ccf5ebef61998aaab5ec6eace8f4564d}
\strng{authornamehash}{ccf5ebef61998aaab5ec6eace8f4564d}
\strng{authorfullhash}{ccf5ebef61998aaab5ec6eace8f4564d}
\field{sortinit}{B}
\field{sortinithash}{8de16967003c7207dae369d874f1456e}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{arXiv preprint arXiv:1409.0473}
\field{title}{Neural machine translation by jointly learning to align and translate}
\field{year}{2014}
\endentry
\entry{bengio2003neural}{article}{}
\name{author}{4}{}{%
{{hash=40a8e4774982146adc2688546f54efb2}{%
family={Bengio},
familyi={B\bibinitperiod},
given={Yoshua},
giveni={Y\bibinitperiod}}}%
{{hash=dcb65deeb8d2de03d00f0aeeb3499771}{%
family={Ducharme},
familyi={D\bibinitperiod},
given={R{é}jean},
giveni={R\bibinitperiod}}}%
{{hash=da21e966c02c3cfd33d74369c7435c1a}{%
family={Vincent},
familyi={V\bibinitperiod},
given={Pascal},
giveni={P\bibinitperiod}}}%
{{hash=bd9126e06687017f43910a8eb056bb0a}{%
family={Jauvin},
familyi={J\bibinitperiod},
given={Christian},
giveni={C\bibinitperiod}}}%
}
\strng{namehash}{058f6aa7282521dda725a309bd2d3fea}
\strng{fullhash}{214c0985c7c874b6fb51e84e6ea66db3}
\strng{bibnamehash}{058f6aa7282521dda725a309bd2d3fea}
\strng{authorbibnamehash}{058f6aa7282521dda725a309bd2d3fea}
\strng{authornamehash}{058f6aa7282521dda725a309bd2d3fea}
\strng{authorfullhash}{214c0985c7c874b6fb51e84e6ea66db3}
\field{sortinit}{B}
\field{sortinithash}{8de16967003c7207dae369d874f1456e}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Journal of machine learning research}
\field{number}{Feb}
\field{title}{A neural probabilistic language model}
\field{volume}{3}
\field{year}{2003}
\field{pages}{1137\bibrangedash 1155}
\range{pages}{19}
\endentry
\entry{brown1990statistical}{article}{}
\name{author}{8}{}{%
{{hash=b269d63ed0974176f251e8034564855e}{%
family={Brown},
familyi={B\bibinitperiod},
given={Peter\bibnamedelima F},
giveni={P\bibinitperiod\bibinitdelim F\bibinitperiod}}}%
{{hash=7e6f41b987a6ec2b9b01a3d33cdd2ddb}{%
family={Cocke},
familyi={C\bibinitperiod},
given={John},
giveni={J\bibinitperiod}}}%
{{hash=4a93f8dffb0593ae148a8ef335eabe24}{%
family={Della\bibnamedelima Pietra},
familyi={D\bibinitperiod\bibinitdelim P\bibinitperiod},
given={Stephen\bibnamedelima A},
giveni={S\bibinitperiod\bibinitdelim A\bibinitperiod}}}%
{{hash=4bbdbda1345838fb46c88bd6afda258c}{%
family={Della\bibnamedelima Pietra},
familyi={D\bibinitperiod\bibinitdelim P\bibinitperiod},
given={Vincent\bibnamedelima J},
giveni={V\bibinitperiod\bibinitdelim J\bibinitperiod}}}%
{{hash=48d2d6e25183229596dbf23cf956498e}{%
family={Jelinek},
familyi={J\bibinitperiod},
given={Frederick},
giveni={F\bibinitperiod}}}%
{{hash=88426f033c21f91cad168597555d363d}{%
family={Lafferty},
familyi={L\bibinitperiod},
given={John},
giveni={J\bibinitperiod}}}%
{{hash=619872e7225f85d3b1c89906ec54e067}{%
family={Mercer},
familyi={M\bibinitperiod},
given={Robert\bibnamedelima L},
giveni={R\bibinitperiod\bibinitdelim L\bibinitperiod}}}%
{{hash=60b30df74a8e4cd8b96789ea5dae645e}{%
family={Roossin},
familyi={R\bibinitperiod},
given={Paul\bibnamedelima S},
giveni={P\bibinitperiod\bibinitdelim S\bibinitperiod}}}%
}
\strng{namehash}{e1a7aeb09589fa95d5b84765dc9acb1d}
\strng{fullhash}{16f3c666fcb85cefe7854f3250ea6ff5}
\strng{bibnamehash}{e1a7aeb09589fa95d5b84765dc9acb1d}
\strng{authorbibnamehash}{e1a7aeb09589fa95d5b84765dc9acb1d}
\strng{authornamehash}{e1a7aeb09589fa95d5b84765dc9acb1d}
\strng{authorfullhash}{16f3c666fcb85cefe7854f3250ea6ff5}
\field{extraname}{1}
\field{sortinit}{B}
\field{sortinithash}{8de16967003c7207dae369d874f1456e}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Computational linguistics}
\field{number}{2}
\field{title}{A statistical approach to machine translation}
\field{volume}{16}
\field{year}{1990}
\field{pages}{79\bibrangedash 85}
\range{pages}{7}
\endentry
\entry{brown1993mathematics}{article}{}
\name{author}{4}{}{%
{{hash=b269d63ed0974176f251e8034564855e}{%
family={Brown},
familyi={B\bibinitperiod},
given={Peter\bibnamedelima F},
giveni={P\bibinitperiod\bibinitdelim F\bibinitperiod}}}%
{{hash=61b61a721b25ebe967463c2b306da5f7}{%
family={Pietra},
familyi={P\bibinitperiod},
given={Vincent\bibnamedelimb J\bibnamedelima Della},
giveni={V\bibinitperiod\bibinitdelim J\bibinitperiod\bibinitdelim D\bibinitperiod}}}%
{{hash=c69a90bf7107509a0bbd82d04ca2a8ae}{%
family={Pietra},
familyi={P\bibinitperiod},
given={Stephen\bibnamedelimb A\bibnamedelima Della},
giveni={S\bibinitperiod\bibinitdelim A\bibinitperiod\bibinitdelim D\bibinitperiod}}}%
{{hash=619872e7225f85d3b1c89906ec54e067}{%
family={Mercer},
familyi={M\bibinitperiod},
given={Robert\bibnamedelima L},
giveni={R\bibinitperiod\bibinitdelim L\bibinitperiod}}}%
}
\list{publisher}{1}{%
{MIT Press}%
}
\strng{namehash}{e1a7aeb09589fa95d5b84765dc9acb1d}
\strng{fullhash}{c933151f7ef6100fa53e2d864a3c0f91}
\strng{bibnamehash}{e1a7aeb09589fa95d5b84765dc9acb1d}
\strng{authorbibnamehash}{e1a7aeb09589fa95d5b84765dc9acb1d}
\strng{authornamehash}{e1a7aeb09589fa95d5b84765dc9acb1d}
\strng{authorfullhash}{c933151f7ef6100fa53e2d864a3c0f91}
\field{extraname}{2}
\field{sortinit}{B}
\field{sortinithash}{8de16967003c7207dae369d874f1456e}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Computational linguistics}
\field{number}{2}
\field{title}{The mathematics of statistical machine translation: Parameter estimation}
\field{volume}{19}
\field{year}{1993}
\field{pages}{263\bibrangedash 311}
\range{pages}{49}
\endentry
\entry{chen2014fast}{inproceedings}{}
\name{author}{2}{}{%
{{hash=aed188db3fc6855ac5666aa14e04a3bb}{%
family={Chen},
familyi={C\bibinitperiod},
given={Danqi},
giveni={D\bibinitperiod}}}%
{{hash=2214edb8305f7ccd7cdc310b3a8ae1b4}{%
family={Manning},
familyi={M\bibinitperiod},
given={Christopher\bibnamedelima D},
giveni={C\bibinitperiod\bibinitdelim D\bibinitperiod}}}%
}
\strng{namehash}{ee9c66797e7327b2ee7b96862e634853}
\strng{fullhash}{ee9c66797e7327b2ee7b96862e634853}
\strng{bibnamehash}{ee9c66797e7327b2ee7b96862e634853}
\strng{authorbibnamehash}{ee9c66797e7327b2ee7b96862e634853}
\strng{authornamehash}{ee9c66797e7327b2ee7b96862e634853}
\strng{authorfullhash}{ee9c66797e7327b2ee7b96862e634853}
\field{sortinit}{C}
\field{sortinithash}{4c244ceae61406cdc0cc2ce1cb1ff703}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{booktitle}{Proceedings of the 2014 conference on empirical methods in natural language processing (EMNLP)}
\field{title}{A fast and accurate dependency parser using neural networks}
\field{year}{2014}
\field{pages}{740\bibrangedash 750}
\range{pages}{11}
\endentry
\entry{chen1999empirical}{article}{}
\name{author}{2}{}{%
{{hash=22329e4fad1d1d229e5afbe4fe5bcca6}{%
family={Chen},
familyi={C\bibinitperiod},
given={Stanley\bibnamedelima F},
giveni={S\bibinitperiod\bibinitdelim F\bibinitperiod}}}%
{{hash=cff3ac3b4d16dde2f682dce454449cff}{%
family={Goodman},
familyi={G\bibinitperiod},
given={Joshua},
giveni={J\bibinitperiod}}}%
}
\list{publisher}{1}{%
{Elsevier}%
}
\strng{namehash}{e7766dbeb1e894d1a1c85e82dc9d26a7}
\strng{fullhash}{e7766dbeb1e894d1a1c85e82dc9d26a7}
\strng{bibnamehash}{e7766dbeb1e894d1a1c85e82dc9d26a7}
\strng{authorbibnamehash}{e7766dbeb1e894d1a1c85e82dc9d26a7}
\strng{authornamehash}{e7766dbeb1e894d1a1c85e82dc9d26a7}
\strng{authorfullhash}{e7766dbeb1e894d1a1c85e82dc9d26a7}
\field{sortinit}{C}
\field{sortinithash}{4c244ceae61406cdc0cc2ce1cb1ff703}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Computer Speech \& Language}
\field{number}{4}
\field{title}{An empirical study of smoothing techniques for language modeling}
\field{volume}{13}
\field{year}{1999}
\field{pages}{359\bibrangedash 394}
\range{pages}{36}
\endentry
\entry{chomsky2002syntactic}{book}{}
\name{author}{1}{}{%
{{hash=28167c0b3800bc5cfb686676277005b0}{%
family={Chomsky},
familyi={C\bibinitperiod},
given={Noam},
giveni={N\bibinitperiod}}}%
}
\list{publisher}{1}{%
{Walter de Gruyter}%
}
\strng{namehash}{28167c0b3800bc5cfb686676277005b0}
\strng{fullhash}{28167c0b3800bc5cfb686676277005b0}
\strng{bibnamehash}{28167c0b3800bc5cfb686676277005b0}
\strng{authorbibnamehash}{28167c0b3800bc5cfb686676277005b0}
\strng{authornamehash}{28167c0b3800bc5cfb686676277005b0}
\strng{authorfullhash}{28167c0b3800bc5cfb686676277005b0}
\field{sortinit}{C}
\field{sortinithash}{4c244ceae61406cdc0cc2ce1cb1ff703}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{Syntactic structures}
\field{year}{2002}
\endentry
\entry{gale1995good}{article}{}
\name{author}{2}{}{%
{{hash=2347dfd9dcf20ea2a38d78f8e6c8e8ab}{%
family={Gale},
familyi={G\bibinitperiod},
given={William\bibnamedelima A},
giveni={W\bibinitperiod\bibinitdelim A\bibinitperiod}}}%
{{hash=cf4650e11caf59af25efb68329bcd096}{%
family={Sampson},
familyi={S\bibinitperiod},
given={Geoffrey},
giveni={G\bibinitperiod}}}%
}
\list{publisher}{1}{%
{Taylor \& Francis}%
}
\strng{namehash}{2ac04dc08f08d1c00b1f04f417ac73bb}
\strng{fullhash}{2ac04dc08f08d1c00b1f04f417ac73bb}
\strng{bibnamehash}{2ac04dc08f08d1c00b1f04f417ac73bb}
\strng{authorbibnamehash}{2ac04dc08f08d1c00b1f04f417ac73bb}
\strng{authornamehash}{2ac04dc08f08d1c00b1f04f417ac73bb}
\strng{authorfullhash}{2ac04dc08f08d1c00b1f04f417ac73bb}
\field{sortinit}{G}
\field{sortinithash}{62eb2aa29549e4fdbd3cb154ec5711cb}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Journal of quantitative linguistics}
\field{number}{3}
\field{title}{Good-turing frequency estimation without tears}
\field{volume}{2}
\field{year}{1995}
\field{pages}{217\bibrangedash 237}
\range{pages}{21}
\endentry
\entry{good1953population}{article}{}
\name{author}{1}{}{%
{{hash=287d039fb19c46dd90dd18326ebee81a}{%
family={Good},
familyi={G\bibinitperiod},
given={Irving\bibnamedelima J},
giveni={I\bibinitperiod\bibinitdelim J\bibinitperiod}}}%
}
\list{publisher}{1}{%
{Oxford University Press}%
}
\strng{namehash}{287d039fb19c46dd90dd18326ebee81a}
\strng{fullhash}{287d039fb19c46dd90dd18326ebee81a}
\strng{bibnamehash}{287d039fb19c46dd90dd18326ebee81a}
\strng{authorbibnamehash}{287d039fb19c46dd90dd18326ebee81a}
\strng{authornamehash}{287d039fb19c46dd90dd18326ebee81a}
\strng{authorfullhash}{287d039fb19c46dd90dd18326ebee81a}
\field{sortinit}{G}
\field{sortinithash}{62eb2aa29549e4fdbd3cb154ec5711cb}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Biometrika}
\field{number}{3-4}
\field{title}{The population frequencies of species and the estimation of population parameters}
\field{volume}{40}
\field{year}{1953}
\field{pages}{237\bibrangedash 264}
\range{pages}{28}
\endentry
\entry{huang2008advanced}{inproceedings}{}
\name{author}{1}{}{%
{{hash=2ba9f491620cf8f3ee610ef7fdf2c9dd}{%
family={Huang},
familyi={H\bibinitperiod},
given={Liang},
giveni={L\bibinitperiod}}}%
}
\strng{namehash}{2ba9f491620cf8f3ee610ef7fdf2c9dd}
\strng{fullhash}{2ba9f491620cf8f3ee610ef7fdf2c9dd}
\strng{bibnamehash}{2ba9f491620cf8f3ee610ef7fdf2c9dd}
\strng{authorbibnamehash}{2ba9f491620cf8f3ee610ef7fdf2c9dd}
\strng{authornamehash}{2ba9f491620cf8f3ee610ef7fdf2c9dd}
\strng{authorfullhash}{2ba9f491620cf8f3ee610ef7fdf2c9dd}
\field{sortinit}{H}
\field{sortinithash}{6db6145dae8dc9e1271a8d556090b50a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{Advanced Dynamic Programming in CL}
\field{year}{2008}
\endentry
\entry{huang2015bidirectional}{article}{}
\name{author}{3}{}{%
{{hash=6e8d947dd72de23b8500095b595e1e99}{%
family={Huang},
familyi={H\bibinitperiod},
given={Zhiheng},
giveni={Z\bibinitperiod}}}%
{{hash=b1dd1a9ff59cc6aeba590fc68bcc39cf}{%
family={Xu},
familyi={X\bibinitperiod},
given={Wei},
giveni={W\bibinitperiod}}}%
{{hash=f5a13dcc79cf0ffe67e50260860f709a}{%
family={Yu},
familyi={Y\bibinitperiod},
given={Kai},
giveni={K\bibinitperiod}}}%
}
\strng{namehash}{2f75424f8afa62a5d143e289d08f1665}
\strng{fullhash}{2f75424f8afa62a5d143e289d08f1665}
\strng{bibnamehash}{2f75424f8afa62a5d143e289d08f1665}
\strng{authorbibnamehash}{2f75424f8afa62a5d143e289d08f1665}
\strng{authornamehash}{2f75424f8afa62a5d143e289d08f1665}
\strng{authorfullhash}{2f75424f8afa62a5d143e289d08f1665}
\field{sortinit}{H}
\field{sortinithash}{6db6145dae8dc9e1271a8d556090b50a}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{arXiv preprint arXiv:1508.01991}
\field{title}{Bidirectional LSTM-CRF models for sequence tagging}
\field{year}{2015}
\endentry
\entry{kneser1995improved}{inproceedings}{}
\name{author}{2}{}{%
{{hash=d9f3c3457f75f60955fa5c2b36fc86ec}{%
family={Kneser},
familyi={K\bibinitperiod},
given={Reinhard},
giveni={R\bibinitperiod}}}%
{{hash=cf68a4c2b64db77cc898cdc9fbdeb0c4}{%
family={Ney},
familyi={N\bibinitperiod},
given={Hermann},
giveni={H\bibinitperiod}}}%
}
\list{organization}{1}{%
{IEEE}%
}
\strng{namehash}{68602597fa7a823464f42d5ef1f755d3}
\strng{fullhash}{68602597fa7a823464f42d5ef1f755d3}
\strng{bibnamehash}{68602597fa7a823464f42d5ef1f755d3}
\strng{authorbibnamehash}{68602597fa7a823464f42d5ef1f755d3}
\strng{authornamehash}{68602597fa7a823464f42d5ef1f755d3}
\strng{authorfullhash}{68602597fa7a823464f42d5ef1f755d3}
\field{sortinit}{K}
\field{sortinithash}{d3edc18d54b9438a72c24c925bfb38f4}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{booktitle}{1995 International Conference on Acoustics, Speech, and Signal Processing}
\field{title}{Improved backing-off for m-gram language modeling}
\field{volume}{1}
\field{year}{1995}
\field{pages}{181\bibrangedash 184}
\range{pages}{4}
\endentry
\entry{koehn2003statistical}{inproceedings}{}
\name{author}{3}{}{%
{{hash=dac0b41a5ea5e261d00f3251b76b91bd}{%
family={Koehn},
familyi={K\bibinitperiod},
given={Philipp},
giveni={P\bibinitperiod}}}%
{{hash=32b0f992581826baf8262f7355351480}{%
family={Och},
familyi={O\bibinitperiod},
given={Franz\bibnamedelima Josef},
giveni={F\bibinitperiod\bibinitdelim J\bibinitperiod}}}%
{{hash=e2a39610bfa3753b6cd9117b5f1c4854}{%
family={Marcu},
familyi={M\bibinitperiod},
given={Daniel},
giveni={D\bibinitperiod}}}%
}
\list{organization}{1}{%
{Association for Computational Linguistics}%
}
\strng{namehash}{9f2b4690ea540aa758f18faa6477fd3a}
\strng{fullhash}{9f2b4690ea540aa758f18faa6477fd3a}
\strng{bibnamehash}{9f2b4690ea540aa758f18faa6477fd3a}
\strng{authorbibnamehash}{9f2b4690ea540aa758f18faa6477fd3a}
\strng{authornamehash}{9f2b4690ea540aa758f18faa6477fd3a}
\strng{authorfullhash}{9f2b4690ea540aa758f18faa6477fd3a}
\field{sortinit}{K}
\field{sortinithash}{d3edc18d54b9438a72c24c925bfb38f4}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{booktitle}{Proceedings of the 2003 Conference of the North American Chapter of the Association for Computational Linguistics on Human Language Technology-Volume 1}
\field{title}{Statistical phrase-based translation}
\field{year}{2003}
\field{pages}{48\bibrangedash 54}
\range{pages}{7}
\endentry
\entry{kolmogorov2018foundations}{book}{}
\name{author}{2}{}{%
{{hash=9d75b8e6905d1c3bbd7c677597929d6c}{%
family={Kolmogorov},
familyi={K\bibinitperiod},
given={Andre\bibnamedelima Nikolaevich},
giveni={A\bibinitperiod\bibinitdelim N\bibinitperiod}}}%
{{hash=528750b676e83a30cc4ff893ee954b75}{%
family={Bharucha-Reid},
familyi={B\bibinithyphendelim R\bibinitperiod},
given={Albert\bibnamedelima T},
giveni={A\bibinitperiod\bibinitdelim T\bibinitperiod}}}%
}
\list{publisher}{1}{%
{Courier Dover Publications}%
}
\strng{namehash}{f43d1e88513d70674f8e142337d15d17}
\strng{fullhash}{f43d1e88513d70674f8e142337d15d17}
\strng{bibnamehash}{f43d1e88513d70674f8e142337d15d17}
\strng{authorbibnamehash}{f43d1e88513d70674f8e142337d15d17}
\strng{authornamehash}{f43d1e88513d70674f8e142337d15d17}
\strng{authorfullhash}{f43d1e88513d70674f8e142337d15d17}
\field{sortinit}{K}
\field{sortinithash}{d3edc18d54b9438a72c24c925bfb38f4}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{Foundations of the theory of probability: Second English Edition}
\field{year}{2018}
\endentry
\entry{lafferty2001conditional}{article}{}
\name{author}{3}{}{%
{{hash=88426f033c21f91cad168597555d363d}{%
family={Lafferty},
familyi={L\bibinitperiod},
given={John},
giveni={J\bibinitperiod}}}%
{{hash=17d73a3a5be48993791cbe4db8855331}{%
family={McCallum},
familyi={M\bibinitperiod},
given={Andrew},
giveni={A\bibinitperiod}}}%
{{hash=b99bb6fa4fedf907a397363d63528047}{%
family={Pereira},
familyi={P\bibinitperiod},
given={Fernando\bibnamedelima CN},
giveni={F\bibinitperiod\bibinitdelim C\bibinitperiod}}}%
}
\strng{namehash}{f1fb6bc84c219f4a487f94edf3d76b14}
\strng{fullhash}{f1fb6bc84c219f4a487f94edf3d76b14}
\strng{bibnamehash}{f1fb6bc84c219f4a487f94edf3d76b14}
\strng{authorbibnamehash}{f1fb6bc84c219f4a487f94edf3d76b14}
\strng{authornamehash}{f1fb6bc84c219f4a487f94edf3d76b14}
\strng{authorfullhash}{f1fb6bc84c219f4a487f94edf3d76b14}
\field{sortinit}{L}
\field{sortinithash}{dad3efd0836470093a7b4a7bb756eb8c}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{Conditional random fields: Probabilistic models for segmenting and labeling sequence data}
\field{year}{2001}
\endentry
\entry{ma2016end}{article}{}
\name{author}{2}{}{%
{{hash=a8f9a3f746656cec93579a6eb8fcd518}{%
family={Ma},
familyi={M\bibinitperiod},
given={Xuezhe},
giveni={X\bibinitperiod}}}%
{{hash=4c40167d6c8ae2073badfccef0c76767}{%
family={Hovy},
familyi={H\bibinitperiod},
given={Eduard},
giveni={E\bibinitperiod}}}%
}
\strng{namehash}{19c7a6df63f9822d4097371fd8d8a305}
\strng{fullhash}{19c7a6df63f9822d4097371fd8d8a305}
\strng{bibnamehash}{19c7a6df63f9822d4097371fd8d8a305}
\strng{authorbibnamehash}{19c7a6df63f9822d4097371fd8d8a305}
\strng{authornamehash}{19c7a6df63f9822d4097371fd8d8a305}
\strng{authorfullhash}{19c7a6df63f9822d4097371fd8d8a305}
\field{sortinit}{M}
\field{sortinithash}{2e5c2f51f7fa2d957f3206819bf86dc3}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{arXiv preprint arXiv:1603.01354}
\field{title}{End-to-end sequence labeling via bi-directional lstm-cnns-crf}
\field{year}{2016}
\endentry
\entry{mikolov2010recurrent}{inproceedings}{}
\name{author}{5}{}{%
{{hash=4a0e59cc1dbb8ca9e8342475f60e326a}{%
family={Mikolov},
familyi={M\bibinitperiod},
given={Tom{á}{š}},
giveni={T\bibinitperiod}}}%
{{hash=532398c21f4c14f31d6c3f5c9fb97d3b}{%
family={Karafi{á}t},
familyi={K\bibinitperiod},
given={Martin},
giveni={M\bibinitperiod}}}%
{{hash=ec089170ef346616c7b4c55ff60fd6cc}{%
family={Burget},
familyi={B\bibinitperiod},
given={Luk{á}{š}},
giveni={L\bibinitperiod}}}%
{{hash=c2b4fbf4b66e87721a2fbe956a495a78}{%
family={{Č}ernock{ỳ}},
familyi={Č\bibinitperiod},
given={Jan},
giveni={J\bibinitperiod}}}%
{{hash=d787f09b6c6cc374a9ad21fe9120fc1f}{%
family={Khudanpur},
familyi={K\bibinitperiod},
given={Sanjeev},
giveni={S\bibinitperiod}}}%
}
\strng{namehash}{2db95ca456b29eae853859553b1267c8}
\strng{fullhash}{8ba380599995fb34b87426fbc79dbda6}
\strng{bibnamehash}{2db95ca456b29eae853859553b1267c8}
\strng{authorbibnamehash}{2db95ca456b29eae853859553b1267c8}
\strng{authornamehash}{2db95ca456b29eae853859553b1267c8}
\strng{authorfullhash}{8ba380599995fb34b87426fbc79dbda6}
\field{sortinit}{M}
\field{sortinithash}{2e5c2f51f7fa2d957f3206819bf86dc3}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{booktitle}{Eleventh annual conference of the international speech communication association}
\field{title}{Recurrent neural network based language model}
\field{year}{2010}
\endentry
\entry{nagao1984framework}{article}{}
\name{author}{1}{}{%
{{hash=a9664468e1992444282044b192539ec1}{%
family={Nagao},
familyi={N\bibinitperiod},
given={Makoto},
giveni={M\bibinitperiod}}}%
}
\strng{namehash}{a9664468e1992444282044b192539ec1}
\strng{fullhash}{a9664468e1992444282044b192539ec1}
\strng{bibnamehash}{a9664468e1992444282044b192539ec1}
\strng{authorbibnamehash}{a9664468e1992444282044b192539ec1}
\strng{authornamehash}{a9664468e1992444282044b192539ec1}
\strng{authorfullhash}{a9664468e1992444282044b192539ec1}
\field{sortinit}{N}
\field{sortinithash}{98cf339a479c0454fe09153a08675a15}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Artificial and human intelligence}
\field{title}{A framework of a mechanical translation between Japanese and English by analogy principle}
\field{year}{1984}
\field{pages}{351\bibrangedash 354}
\range{pages}{4}
\endentry
\entry{ney1994structuring}{article}{}
\name{author}{3}{}{%
{{hash=cf68a4c2b64db77cc898cdc9fbdeb0c4}{%
family={Ney},
familyi={N\bibinitperiod},
given={Hermann},
giveni={H\bibinitperiod}}}%
{{hash=b2ef0258c04f3ad7e91e09d0279e802e}{%
family={Essen},
familyi={E\bibinitperiod},
given={Ute},
giveni={U\bibinitperiod}}}%
{{hash=d9f3c3457f75f60955fa5c2b36fc86ec}{%
family={Kneser},
familyi={K\bibinitperiod},
given={Reinhard},
giveni={R\bibinitperiod}}}%
}
\strng{namehash}{55fd427c3f707df2bb830dcbeac8cfeb}
\strng{fullhash}{55fd427c3f707df2bb830dcbeac8cfeb}
\strng{bibnamehash}{55fd427c3f707df2bb830dcbeac8cfeb}
\strng{authorbibnamehash}{55fd427c3f707df2bb830dcbeac8cfeb}
\strng{authornamehash}{55fd427c3f707df2bb830dcbeac8cfeb}
\strng{authorfullhash}{55fd427c3f707df2bb830dcbeac8cfeb}
\field{sortinit}{N}
\field{sortinithash}{98cf339a479c0454fe09153a08675a15}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Computer Speech \& Language}
\field{number}{1}
\field{title}{On structuring probabilistic dependences in stochastic language modelling}
\field{volume}{8}
\field{year}{1994}
\field{pages}{1\bibrangedash 38}
\range{pages}{38}
\endentry
\entry{ng2002discriminative}{inproceedings}{}
\name{author}{2}{}{%
{{hash=49e889356ff39df159461bc2895c7e16}{%
family={Ng},
familyi={N\bibinitperiod},
given={Andrew\bibnamedelima Y},
giveni={A\bibinitperiod\bibinitdelim Y\bibinitperiod}}}%
{{hash=8a36116840c7ee55901618c95fd08a58}{%
family={Jordan},
familyi={J\bibinitperiod},
given={Michael\bibnamedelima I},
giveni={M\bibinitperiod\bibinitdelim I\bibinitperiod}}}%
}
\strng{namehash}{dca981b2891288e8bdd7d6fbf1a6a9c4}
\strng{fullhash}{dca981b2891288e8bdd7d6fbf1a6a9c4}
\strng{bibnamehash}{dca981b2891288e8bdd7d6fbf1a6a9c4}
\strng{authorbibnamehash}{dca981b2891288e8bdd7d6fbf1a6a9c4}
\strng{authornamehash}{dca981b2891288e8bdd7d6fbf1a6a9c4}
\strng{authorfullhash}{dca981b2891288e8bdd7d6fbf1a6a9c4}
\field{sortinit}{N}
\field{sortinithash}{98cf339a479c0454fe09153a08675a15}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{booktitle}{Advances in neural information processing systems}
\field{title}{On discriminative vs. generative classifiers: A comparison of logistic regression and naive bayes}
\field{year}{2002}
\field{pages}{841\bibrangedash 848}
\range{pages}{8}
\endentry
\entry{papineni2002bleu}{inproceedings}{}
\name{author}{4}{}{%
{{hash=7f72c67734e28d79452d687f2b2825d0}{%
family={Papineni},
familyi={P\bibinitperiod},
given={Kishore},
giveni={K\bibinitperiod}}}%
{{hash=fcdf7cf3f3cfe87d0f08f8f4d98b1137}{%
family={Roukos},
familyi={R\bibinitperiod},
given={Salim},
giveni={S\bibinitperiod}}}%
{{hash=630f7af400534d0e9f11cc0befbb3fe3}{%
family={Ward},
familyi={W\bibinitperiod},
given={Todd},
giveni={T\bibinitperiod}}}%
{{hash=ecc966c09b4816bda3be0efb9bedb125}{%
family={Zhu},
familyi={Z\bibinitperiod},
given={Wei-Jing},
giveni={W\bibinithyphendelim J\bibinitperiod}}}%
}
\list{organization}{1}{%
{Association for Computational Linguistics}%
}
\strng{namehash}{d4eb40956a6de2c2fe50bfef6a72df63}
\strng{fullhash}{0a6e6b36539eaebe00a3fef93d4cd32a}
\strng{bibnamehash}{d4eb40956a6de2c2fe50bfef6a72df63}
\strng{authorbibnamehash}{d4eb40956a6de2c2fe50bfef6a72df63}
\strng{authornamehash}{d4eb40956a6de2c2fe50bfef6a72df63}
\strng{authorfullhash}{0a6e6b36539eaebe00a3fef93d4cd32a}
\field{sortinit}{P}
\field{sortinithash}{bb5b15f2db90f7aef79bb9e83defefcb}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{booktitle}{Proceedings of the 40th annual meeting on association for computational linguistics}
\field{title}{BLEU: a method for automatic evaluation of machine translation}
\field{year}{2002}
\field{pages}{311\bibrangedash 318}
\range{pages}{8}
\endentry
\entry{parsing2009speech}{article}{}
\name{author}{1}{}{%
{{hash=39bd56e32c83d8349b276dbb954c1bd6}{%
family={Parsing},
familyi={P\bibinitperiod},
given={Constituency},
giveni={C\bibinitperiod}}}%
}
\strng{namehash}{39bd56e32c83d8349b276dbb954c1bd6}
\strng{fullhash}{39bd56e32c83d8349b276dbb954c1bd6}
\strng{bibnamehash}{39bd56e32c83d8349b276dbb954c1bd6}
\strng{authorbibnamehash}{39bd56e32c83d8349b276dbb954c1bd6}
\strng{authornamehash}{39bd56e32c83d8349b276dbb954c1bd6}
\strng{authorfullhash}{39bd56e32c83d8349b276dbb954c1bd6}
\field{sortinit}{P}
\field{sortinithash}{bb5b15f2db90f7aef79bb9e83defefcb}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{Speech and language processing}
\field{year}{2009}
\endentry
\entry{resnick1992adventures}{book}{}
\name{author}{1}{}{%
{{hash=2becf933ecad0836916af8d8e92ca548}{%
family={Resnick},
familyi={R\bibinitperiod},
given={Sidney\bibnamedelima I},
giveni={S\bibinitperiod\bibinitdelim I\bibinitperiod}}}%
}
\list{publisher}{1}{%
{Springer Science \& Business Media}%
}
\strng{namehash}{2becf933ecad0836916af8d8e92ca548}
\strng{fullhash}{2becf933ecad0836916af8d8e92ca548}
\strng{bibnamehash}{2becf933ecad0836916af8d8e92ca548}
\strng{authorbibnamehash}{2becf933ecad0836916af8d8e92ca548}
\strng{authornamehash}{2becf933ecad0836916af8d8e92ca548}
\strng{authorfullhash}{2becf933ecad0836916af8d8e92ca548}
\field{sortinit}{R}
\field{sortinithash}{b9c68a358aea118dfa887b6e902414a7}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{Adventures in stochastic processes}
\field{year}{1992}
\endentry
\entry{shannon1948mathematical}{article}{}
\name{author}{1}{}{%
{{hash=536c91678d1dde76c9acd47047556291}{%
family={Shannon},
familyi={S\bibinitperiod},
given={Claude\bibnamedelima E},
giveni={C\bibinitperiod\bibinitdelim E\bibinitperiod}}}%
}
\list{publisher}{1}{%
{Wiley Online Library}%
}
\strng{namehash}{536c91678d1dde76c9acd47047556291}
\strng{fullhash}{536c91678d1dde76c9acd47047556291}
\strng{bibnamehash}{536c91678d1dde76c9acd47047556291}
\strng{authorbibnamehash}{536c91678d1dde76c9acd47047556291}
\strng{authornamehash}{536c91678d1dde76c9acd47047556291}
\strng{authorfullhash}{536c91678d1dde76c9acd47047556291}
\field{extraname}{1}
\field{sortinit}{S}
\field{sortinithash}{c319cff79d99c853d775f88277d4e45f}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Bell system technical journal}
\field{number}{3}
\field{title}{A mathematical theory of communication}
\field{volume}{27}
\field{year}{1948}
\field{pages}{379\bibrangedash 423}
\range{pages}{45}
\endentry
\entry{shannon1949communication}{article}{}
\name{author}{1}{}{%
{{hash=536c91678d1dde76c9acd47047556291}{%
family={Shannon},
familyi={S\bibinitperiod},
given={Claude\bibnamedelima E},
giveni={C\bibinitperiod\bibinitdelim E\bibinitperiod}}}%
}
\list{publisher}{1}{%
{Wiley Online Library}%
}
\strng{namehash}{536c91678d1dde76c9acd47047556291}
\strng{fullhash}{536c91678d1dde76c9acd47047556291}
\strng{bibnamehash}{536c91678d1dde76c9acd47047556291}
\strng{authorbibnamehash}{536c91678d1dde76c9acd47047556291}
\strng{authornamehash}{536c91678d1dde76c9acd47047556291}
\strng{authorfullhash}{536c91678d1dde76c9acd47047556291}
\field{extraname}{2}
\field{sortinit}{S}
\field{sortinithash}{c319cff79d99c853d775f88277d4e45f}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Bell system technical journal}
\field{number}{4}
\field{title}{Communication theory of secrecy systems}
\field{volume}{28}
\field{year}{1949}
\field{pages}{656\bibrangedash 715}
\range{pages}{60}
\endentry
\entry{shiwen1993automatic}{article}{}
\name{author}{1}{}{%
{{hash=79a954956a47a1a00e0fe51f901be97a}{%
family={Shiwen},
familyi={S\bibinitperiod},
given={Yu},
giveni={Y\bibinitperiod}}}%
}
\list{publisher}{1}{%
{Springer}%
}
\strng{namehash}{79a954956a47a1a00e0fe51f901be97a}
\strng{fullhash}{79a954956a47a1a00e0fe51f901be97a}
\strng{bibnamehash}{79a954956a47a1a00e0fe51f901be97a}
\strng{authorbibnamehash}{79a954956a47a1a00e0fe51f901be97a}
\strng{authornamehash}{79a954956a47a1a00e0fe51f901be97a}
\strng{authorfullhash}{79a954956a47a1a00e0fe51f901be97a}
\field{sortinit}{S}
\field{sortinithash}{c319cff79d99c853d775f88277d4e45f}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{Machine Translation}
\field{number}{1-2}
\field{title}{Automatic evaluation of output quality for machine translation systems}
\field{volume}{8}
\field{year}{1993}
\field{pages}{117\bibrangedash 126}
\range{pages}{10}
\endentry
\entry{snover2006study}{inproceedings}{}
\name{author}{5}{}{%
{{hash=57fddde22dd45c3fd89f698609a15e64}{%
family={Snover},
familyi={S\bibinitperiod},
given={Matthew},
giveni={M\bibinitperiod}}}%
{{hash=c401d97f1db13573221c0d16dc3cf8c8}{%
family={Dorr},
familyi={D\bibinitperiod},
given={Bonnie},
giveni={B\bibinitperiod}}}%
{{hash=ed9b7985941f821b7c600e3973f62e51}{%
family={Schwartz},
familyi={S\bibinitperiod},
given={Richard},
giveni={R\bibinitperiod}}}%
{{hash=0a0cf3ca917fdcd5c1b176d34e1d6a7f}{%
family={Micciulla},
familyi={M\bibinitperiod},
given={Linnea},
giveni={L\bibinitperiod}}}%
{{hash=be711c1d20e2f962368bf225974dc978}{%
family={Makhoul},
familyi={M\bibinitperiod},
given={John},
giveni={J\bibinitperiod}}}%
}
\strng{namehash}{a778f5e3fb96c1f0f14573ab770f17d2}
\strng{fullhash}{3c6143a66e4a7a04971c5c4c84f895fe}
\strng{bibnamehash}{a778f5e3fb96c1f0f14573ab770f17d2}
\strng{authorbibnamehash}{a778f5e3fb96c1f0f14573ab770f17d2}
\strng{authornamehash}{a778f5e3fb96c1f0f14573ab770f17d2}
\strng{authorfullhash}{3c6143a66e4a7a04971c5c4c84f895fe}
\field{sortinit}{S}
\field{sortinithash}{c319cff79d99c853d775f88277d4e45f}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{booktitle}{Proceedings of association for machine translation in the Americas}
\field{number}{6}
\field{title}{A study of translation edit rate with targeted human annotation}
\field{volume}{200}
\field{year}{2006}
\endentry
\entry{sutskever2014sequence}{inproceedings}{}
\name{author}{3}{}{%
{{hash=8d569d1d5b8b5a7836017a98b430f959}{%
family={Sutskever},
familyi={S\bibinitperiod},
given={Ilya},
giveni={I\bibinitperiod}}}%
{{hash=494b568c5dc85ba8f3f409635f9c5f25}{%
family={Vinyals},
familyi={V\bibinitperiod},
given={Oriol},
giveni={O\bibinitperiod}}}%
{{hash=c636f146591d51579a8119b777394878}{%
family={Le},
familyi={L\bibinitperiod},
given={Quoc\bibnamedelima V},
giveni={Q\bibinitperiod\bibinitdelim V\bibinitperiod}}}%
}
\strng{namehash}{d60494862bf48faf76d925f3780a6aee}
\strng{fullhash}{d60494862bf48faf76d925f3780a6aee}
\strng{bibnamehash}{d60494862bf48faf76d925f3780a6aee}
\strng{authorbibnamehash}{d60494862bf48faf76d925f3780a6aee}
\strng{authornamehash}{d60494862bf48faf76d925f3780a6aee}
\strng{authorfullhash}{d60494862bf48faf76d925f3780a6aee}
\field{sortinit}{S}
\field{sortinithash}{c319cff79d99c853d775f88277d4e45f}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{booktitle}{Advances in neural information processing systems}
\field{title}{Sequence to sequence learning with neural networks}
\field{year}{2014}
\field{pages}{3104\bibrangedash 3112}
\range{pages}{9}
\endentry
\entry{zhu2015long}{article}{}
\name{author}{3}{}{%
{{hash=7db206795c9c6671b7060b6153bcdab6}{%
family={Zhu},
familyi={Z\bibinitperiod},
given={Xiaodan},
giveni={X\bibinitperiod}}}%
{{hash=aafed55fdbb7a25c1ab43ed724eb953c}{%
family={Sobhani},
familyi={S\bibinitperiod},
given={Parinaz},
giveni={P\bibinitperiod}}}%
{{hash=9ef1c27eeeb236462f52032de708fe81}{%
family={Guo},
familyi={G\bibinitperiod},
given={Hongyu},
giveni={H\bibinitperiod}}}%
}
\strng{namehash}{08aaf389d8d0218d66adedd3df030f19}
\strng{fullhash}{08aaf389d8d0218d66adedd3df030f19}
\strng{bibnamehash}{08aaf389d8d0218d66adedd3df030f19}
\strng{authorbibnamehash}{08aaf389d8d0218d66adedd3df030f19}
\strng{authornamehash}{08aaf389d8d0218d66adedd3df030f19}
\strng{authorfullhash}{08aaf389d8d0218d66adedd3df030f19}
\field{sortinit}{Z}
\field{sortinithash}{8f7b480688e809b50b6f6577b16f3db5}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{journaltitle}{arXiv: Computation and Language}
\field{title}{Long Short-Term Memory Over Tree Structures.}
\field{year}{2015}
\endentry
\entry{liuke-markov-2004}{book}{}
\name{author}{1}{}{%
{{hash=c01ef9a560378730f00ebeecceb572d4}{%
family={刘克},
familyi={刘\bibinitperiod}}}%
}
\list{publisher}{1}{%
{清华大学出版社}%
}
\strng{namehash}{c01ef9a560378730f00ebeecceb572d4}
\strng{fullhash}{c01ef9a560378730f00ebeecceb572d4}
\strng{bibnamehash}{c01ef9a560378730f00ebeecceb572d4}
\strng{authorbibnamehash}{c01ef9a560378730f00ebeecceb572d4}
\strng{authornamehash}{c01ef9a560378730f00ebeecceb572d4}
\strng{authorfullhash}{c01ef9a560378730f00ebeecceb572d4}
\field{sortinit}{刘}
\field{sortinithash}{f3ff24b163d4496b49484b7a115b76c9}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{实用马尔可夫决策过程}
\field{volume}{3}
\field{year}{2004}
\endentry
\entry{mao-prob-book-2011}{book}{}
\name{author}{6}{}{%
{{hash=2830831f67df549d8bcc4980e281a6ca}{%
family={茆诗松},
familyi={茆\bibinitperiod}}}%
{{hash=07f329c27730232bd243800625979283}{%
family={程依明},
familyi={程\bibinitperiod}}}%
{{hash=614b04f1697449a3d7a6615f9552d64d}{%
family={濮晓龙},
familyi={濮\bibinitperiod}}}%
{{hash=75212c77695d1b78aaf6d5f383728eb4}{%
family={平装},
familyi={平\bibinitperiod}}}%
{{hash=4d7ffbb58d61f76765dea25a0d4da1ec}{%
family={查看清},
familyi={查\bibinitperiod}}}%
{{hash=562a92d3f2b96306488b04b8b94b427a}{%
family={单书目},
familyi={单\bibinitperiod}}}%
}
\list{publisher}{1}{%
{北京: 高等教育出版社}%
}
\strng{namehash}{a5d297d55af03f5c30541752e4e66972}
\strng{fullhash}{0cd8944e1838d053c86a350a486cd430}
\strng{bibnamehash}{a5d297d55af03f5c30541752e4e66972}
\strng{authorbibnamehash}{a5d297d55af03f5c30541752e4e66972}
\strng{authornamehash}{a5d297d55af03f5c30541752e4e66972}
\strng{authorfullhash}{0cd8944e1838d053c86a350a486cd430}
\field{sortinit}{茆}
\field{sortinithash}{e6b0878d3262ba263a8d6ee4986b856f}
\field{labelnamesource}{author}
\field{labeltitlesource}{title}
\field{title}{概率论与数理统计教程: 第二版}
\field{year}{2011}
\endentry
\enddatalist
\endrefsection
\endinput
Book/mt-book.tex
......@@ -102,8 +102,8 @@
%----------------------------------------------------------------------------------------
% CHAPTERS
%----------------------------------------------------------------------------------------
\include{Chapter1/chapter1}
\include{Chapter2/chapter2}
%\include{Chapter1/chapter1}
%\include{Chapter2/chapter2}
\include{Chapter3/chapter3}
......@@ -118,7 +118,7 @@
%\chapter*{Bibliography}
%\addcontentsline{toc}{chapter}{\textcolor{ocre}{Bibliography}} % 增加一个标题
%\printbibliography[heading=bibempty]
%\printbibliography
\printbibliography
%------------------------------------------------
......
Book/structure.tex
......@@ -544,7 +544,7 @@ addtohook={%
\usepackage[justification=centering]{caption}%强制图片居中
\usepackage{subfigure}
\newcommand{\parinterval}{\noindent\hspace{2em}}%定义变量替代原来开头的控制缩进
\usepackage{tikz-qtree}
......
Markdown 格式
0% 或
您添加了 0 人 到此讨论。请谨慎行事。
请先完成此评论的编辑！
请 注册 或者 后发表评论