DMCA
Sentiment Analyzer for Arabic Comments System
| Citations: | 2 - 0 self |
BibTeX
@MISC{El-dine_sentimentanalyzer,
author = {Alaa El-dine and Ali Hamouda},
title = {Sentiment Analyzer for Arabic Comments System},
year = {}
}
Share
 |
 |
 |
 |
||
OpenURL
Abstract
Abstract—Today, the number of users of social network is increasing. Millions of users share opinions on different aspects of life every day. Therefore social network are rich sources of data for opinion mining and sentiment analysis. Also users have become more interested in following news pages on Facebook. Several posts; political for example, have thousands of users’ comments that agree/disagree with the post content. Such comments can be a good indicator for the community opinion about the post content. For politicians, marketers, decision makers …, it is required to make sentiment analysis to know the percentage of users agree, disagree and neutral respect to a post. This raised the need to analyze theusers ’ comments in Facebook. We focused on Arabic Facebook news pages for the task of sentiment analysis. We developed a corpus for sentiment analysis and opinion mining purposes. Then, we used different machine learning algorithms – decision tree, support vector machines, and naive bayes- to develop sentiment analyzer. The performance of the system using each technique was evaluated and compared with others. Keywords—Analysis for Arabic comments; machine learning algorithms; sentiment analysis; opinion mining I.
Keyphrases
sentiment analysis sentiment analyzer arabic comment system post content opinion mining good indicator machine learning algorithm keywords analysis several post therefore social network community opinion news page rich source support vector machine arabic facebook news page theusers comment abstract today user comment naive bayes social network user share opinion opinion mining purpose neutral respect different machine arabic comment algorithm decision tree decision maker different aspect
