-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathreport.tex
132 lines (106 loc) · 3.43 KB
/
report.tex
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
\documentclass[12pt]{report}
\linespread{2}
\usepackage[pdftex]{graphicx}
\usepackage{amssymb}
\usepackage{tocbibind}
\usepackage[utf8]{inputenc}
\usepackage[english]{babel}
\usepackage{natbib}
\usepackage{amsmath}
\usepackage{multirow}
\usepackage{graphicx}
\usepackage{stackengine}
\usepackage{subcaption}
\usepackage[margin=1.25in]{geometry}
\usepackage[nodayofweek]{datetime}
\newdateformat{mydate}{{ }\shortmonthname[\THEMONTH], \THEYEAR}
\newdateformat{myyear}{{ }\THEYEAR}
% latex
% \usepackage[hidelinks]{hyperref}
\usepackage{hyperref}
\usepackage[hyphenbreaks]{breakurl}
\usepackage{booktabs}
%\usepackage{lineno}
%\linenumbers
\usepackage{calc}
\usepackage{enumitem}
%\usepackage{sectsty}
%
%%\chapternumberfont{\tiny}
%\chaptertitlefont{\huge}
%opening
\title{Improving Neural Machine Translation for Morphologically rich languages }
\author{Raja Gunasekaran}
\usepackage{graphicx}
\usepackage{pdfpages}
\graphicspath{ {images/} }
\begin{document}
\pagenumbering{roman}
%\thispagestyle{empty} %remove page number from title page
\includepdf{tex/title.pdf}
%\setcounter{page}{2} % Title Page gets number "i"
%\thispagestyle{empty} %remove page number from title page
%\input{tex/title.tex}
\pagestyle{plain}
\clearpage
\input{tex/abstract.tex}
\phantomsection\addcontentsline{toc}{chapter}{Abstract}
\tableofcontents
\listoftables
\listoffigures
\newpage
\pagenumbering{arabic}
%\addcontentsline{toc}{chapter}{\abstractname}
%\begin{abstract}
%\end{abstract}
%\cleardoublepage
%\listoffigures
%\cleardoublepage
%\addcontentsline{toc}{chapter}{\listfigurename}
%\listoftables
%\cleardoublepage
%\addcontentsline{toc}{chapter}{\listtablename}
\chapter{Introduction}
\input{tex/intro.tex}
\chapter{Background}
\label{background}
\input{tex/background.tex}
\chapter{Handling Rare and Unknown Words}
\label{related}
\input{tex/related.tex}
\chapter{Vocabulary Expansion for NMT}
\label{proposed}
\input{tex/proposed.tex}
\chapter{Vector Representations for OOV words}
\label{comparision1}
\input{tex/experiment1.tex}
\chapter{MT for Morphologically Rich Languages}
\label{experiments}
\input{tex/experiment2.tex}
\chapter{Conclusion and Future Work}
\label{conclusion}
\input{tex/conclusion.tex}
%\input{tex/timeline.tex}
\newpage
\appendix
\chapter{List of Abbreviations and Definitions}
\begin{description}[leftmargin=!,labelwidth=\widthof{\bfseries The longest label}]
\item[BLEU] Bilingual Evaluation Understudy
\item[BPE] Byte Pair Encoding
\item[CNN] Convolution Neural Network
\item[GRU] Gated Recurrent Unit
\item[LSTM] Long Short-Term Memory
\item[MT] Machine Translation
\item[NMT] Neural Machine Translation
\item[OOV] Out Of Vocabulary words
\item[RNN] Recurrent Neural Network
\item[SMT] Statistical Machine Translation
\item[Morpheme] The smallest meaningful units in a language.
\item[Word embedding] A technique used to map and represent words in a language as vectors of real number, where words with similar meaning have similar representations.
\item[Word vector] A vector used to represent a word in the vector space. Also known as \textit{Word representations}
\item[Morphological rich languages] Languages that encode large amount of information through morphology. They have a very large vocabulary size since there can large number of word forms per lexeme.
\end{description}
% SUmmarize key points. less emphasis on technical grounds. Put in exact technical content.
\bibliography{bib}
\bibliographystyle{apalike}
\end{document}