A comprehensive survey of arabic sentiment analysis

Abstract

Sentiment analysis (SA) is a continuing field of research that lies at the intersection of many fields such as data mining, natural language processing and machine learning. It is concerned with the automatic extraction of opinions conveyed in a certain text. Due to its vast applications, many studies have been conducted in the area of SA especially on English texts, while other languages such as Arabic received less attention. This survey presents a comprehensive overview of the works done so far on Arabic SA (ASA). The survey groups published papers based on the SA-related problems they address and tries to identify the gaps in the current literature laying foundation for future studies in this field.