Arabic Sentiment Analysis Approaches: An Overview

Abstract

In the last few years, Arabic Sentiment Analysis (ASA) has become the most important trends topic of market and scientific research in Machine Learning (ML) and Natural Language Processing (NLP) fields. It tries to identify the opinion orientation (positivity, neutrality or negativity) of a piece text in different domains and it has gained significant results especially in the field of education, in other words ASA is utilized to ameliorate education by trying to understand the learning degree of students and monitoring the performance of students. Due to Arabic complexities at the level of orthography, morphology and dialects. Sentiment analysis for Arabic language is very difficult and it considered as a more challenging process compared to other languages like English. In this work, we tried to give a comparative analysis of Arabic Sentiment approaches, together with various powerful existing techniques.