Sentiment Analysis of Women Driving in Saudi Arabia  
Author Hanan Muhajab

 

Co-Author(s) Kambiz Ghazinour

 

Abstract Twitter is the most popular and valuable resource for sentiment analysis. This observation holds because the platform ensures that individuals express their thoughts and ideas regarding various topics, especially about social life. With users restricted to 140 characters or less, Twitter has evolved as one of the most dominant social networking websites in the world. This research has two main aspects. The first objective is to demonstrate the implementation of sentiment analyzer for Tweeter’s comments, which is considered as one of the most critical and freely big data sources. We analyzed Arabic text using Saudi dialect tweets to get the sentiments around a particular topic in Saudi Arabia. After the preprocessing stage, the collected tweets were analyzed by using two of the machinelearning algorithms: Support Vector Machine and Naïve Bayes. Indeed, we evaluated the algorithms implemented by using confusion matrix. The second objective was to examine the challenges and complexities in Arabic text, especially those that make the Arabic sentiment analysis more complex than other languages.

 

Keywords Sentiment Analysis, Twitter Analysis, Data Mining, Supervised Approach, Weka
   
    Article #:  DSIS19-64
 
Proceedings of ISSAT International Conference on Data Science & Intelligent Systems
August 1-3, 2019 - Las Vegas, NV, U.S.A.