Advertisement
Not a member of Pastebin yet?
Sign Up,
it unlocks many cool features!
- \documentclass[dvips,12pt]{article}
- % Any percent sign marks a comment to the end of the line
- % Every latex document starts with a documentclass declaration like this
- % The option dvips allows for graphics, 12pt is the font size, and article
- % is the style
- \usepackage[pdftex]{graphicx}
- \usepackage{url}
- % These are additional packages for "pdflatex", graphics, and to include
- % hyperlinks inside a document.
- \setlength{\oddsidemargin}{0.25in}
- \setlength{\textwidth}{6.5in}
- \setlength{\topmargin}{0in}
- \setlength{\textheight}{8.5in}
- % These force using more of the margins that is the default style
- \begin{document}
- % Everything after this becomes content
- % Replace the text between curly brackets with your own
- \title{Graduation Project Proposal}
- \author{Ahmed Shawky\\Hassan Mohamed\\MennatAllah Hany\\Mostafa Mohamed\\Moamen Ahmed\\Yusuf Hussien}
- \date{\today}
- % You can leave out "date" and it will be added automatically for today
- % You can change the "\today" date to any text you like
- \maketitle
- % This command causes the title to be created in the document
- \section{Title}
- User Type Classification of Tweets with Implications for Event Recognition in English and Arabic Languages.
- \section{Introduction}
- Twitter has become the social media channel of choice for world leaders as well as organizations to reach large audiences.Consequently, It is beneficial to classify tweets into two user types - organizations and individuals to improve downstream event recognition tasks.
- \section{Problem Statement}
- The goal of this project is to recognize twitter events according to user type in English and Arabic languages.Individual persons tweets tend to report local events unlike organizations tweets which have wider audience. Mainly we aim to classify the tweets according to their textual content only.However, classification may be supported by some complementary sources of information as profile data. There are some features that could help in the text classification process as sentiment and emotion expression.In addition to, the informal language usage, tweet style and news head line.
- \section{Motivation}
- The motive behind this idea, is that recent related studies/papers were made by tackling different languages but those in Arabic are limited, papers showed that the maximum accuracy reached for a specific language ranged approximately between 87 to 89 percent.Thus the objective behind this paper is achieving a high percentage of accuracy by using various advanced techniques and using the Arabic language.
- \section{Data sets}
- The data sets that are going to be used will be derived from :\\
- - Data sets extracted from Twitter API for a suitable period of time\\
- - Data sets collected by some researchers who have worked on a similar research paper\\
- - Data sets that are generated manually by research group members.
- \section{Related Work}
- The same project idea was published in a paper\cite{paper} but the research was done targeting English and Spanish Languages.
- \begin{thebibliography}{99}
- \bibitem{paper}
- Lalindra De Silva and Ellen Riloff
- {User Type Classification of Tweets with Implications for Event
- Recognition}.\\
- \url{http://w...content-available-to-author-only...h.edu/~riloff/pdfs/DeSilva-SMWorkshop-ACL14.pdf}
- \end{thebibliography}
- \end{document}
Advertisement
Add Comment
Please, Sign In to add comment
Advertisement