Skip to content

Latest commit

 

History

History
26 lines (15 loc) · 929 Bytes

README.md

File metadata and controls

26 lines (15 loc) · 929 Bytes

Sentiment-analysis-arabic

This project was part of my research during my Master's thesis.

Title of thesis: Predictive sentiment analysis of tweets using natural language processing techniques.

Case study: Arabic Twittersphere

Test accuracy: 81%

Project: Research on sentiment analysis is used to classify tweets from the Twitter social media platform based on the Hate sentiment. Two techniques were used :

  • Natural Language Processing
  • Machine Learning Classifiers The goal is to Automatically Detect Hate Speech in the Tweets from the Arabic Twittersphere written in the Arabic language, in order to reduce the spread of Hate crimes in social media.

Results

Thesis Advisor: Dr. Lazhar CHINE, Dept of Applied Statistics and econometrics, ENSSEA, Kolea university campus.