Arabic Aspect-Based Sentiment Analysis: A Systematic Literature Review

• Published in: IEEE access Vol. 9; pp. 152628 - 152645
• Main Authors: Obiedat, Ruba, Al-Darras, Duha, Alzaghoul, Esra, Harfoushi, Osama
• Format: Journal Article
• Language: English
• Published: Piscataway IEEE 2021; The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
• Subjects: Arabic sentiment analysis; aspect-based sentiment analysis; Data mining; Datasets; Feature extraction; feature-based sentiment analysis; Hidden Markov models; Integrated circuits; Literature reviews; multi-aspect sentiment analysis; Natural language processing; Sentiment analysis; Systematics; Task analysis; Texts; Websites
• ISSN: 2169-3536; 2169-3536
• DOI: 10.1109/ACCESS.2021.3127140

Recently sentiment analysis in Arabic has attracted much attention from researchers. A modest number of studies have been conducted on Arabic sentiment analysis. However, due to the vast increase in users' comments and reviews on social media and e-commerce websites, the necessity to detect sentence-level and aspect-level sentiments has also increased. The aspect-based sentiment analysis has emerged to detect sentiments at the aspect level. Few studies have attempted to perform aspect-based sentiment analysis on Arabic texts because Arabic natural language processing is a challenging task and because of the lack of available Arabic annotated corpora. In this paper, we conducted a systematic review of the methods, techniques, and datasets employed in aspect-based sentiment analysis on Arabic texts. A total of 21 articles published between 2015-2021 were included in this review. After analysing these articles, we found a lack of annotated datasets that can be used by researchers. In addition, the used datasets were limited to few fields. This review will serve as a foundation for researchers interested in Aspect-Based Sentiment Analysis, it will assist them in developing new models and techniques to tackle this task in the future.