|
" A New Opinion Review Methodology for Arabic Twitter Sentiment Analysis "
Almalki, Abdullah S.
Taylor, Bradley
| Document Type
|
:
|
Latin Dissertation
|
| Language of Document
|
:
|
English
|
| Record Number
|
:
|
1054559
|
| Doc. No
|
:
|
TL53676
|
| Main Entry
|
:
|
Almalki, Abdullah S.
|
| Title & Author
|
:
|
A New Opinion Review Methodology for Arabic Twitter Sentiment Analysis\ Almalki, Abdullah S.Taylor, Bradley
|
| College
|
:
|
The Catholic University of America
|
| Date
|
:
|
2020
|
| Degree
|
:
|
M.S.C.S.
|
| student score
|
:
|
2020
|
| Note
|
:
|
63 p.
|
| Abstract
|
:
|
Each day, a tremendous number of opinions are posted about different issues and different subjects from massive active users on social media. While people increasingly use social media applications such as Twitter to express their feelings and opinions about many things such as events, movies, services and products; this data may also be used as a more valuable resource for governments, companies and decision-makers informing their decisions. This research proposed to use a machine learning approach and Lexicon approach to extract opinions of groups of people from Arabic tweets. In this research I used an entertainment event, as a case study. This research focuses on assessing peoples' opinions more accurately by leveraging the Retweet and Favorite, providing a similar effect as the tweet itself, as well as removing tweets unrelated to the subject opinion, such as: news, advertisements and spam. This research redounds to extend current knowledge of social media analysis and machine learning in the field of opinion analysis. This provides a methodology for selecting effective sentiment analysis mechanisms for Arabic tweets; in particular, accurate results using the developed machine learning algorithm.
|
| Descriptor
|
:
|
Artificial intelligence
|
|
|
:
|
Computer science
|
| Added Entry
|
:
|
Taylor, Bradley
|
| Added Entry
|
:
|
The Catholic University of America
|
| |