Deep Learning Approach in Sentiment Analysis: A Review
Enas A. Hakim Khalil, Enas M. F. El Houby, Hoda Korashy Mohamed · 2020
Sentiment Analysis (SA) is the field that combines Natural Language Processing (NLP), Computational Linguistics (CL) and text analysis to study people's opinions through, by extracting and analyzing subjective information from different resources as the Web, social media and similar sources and so help in drawing public's sentiments or attitude toward certain people, products or ideas and extracting the contextual polarity of the information. This review focuses on recent work in SA using Deep Learning (DL)techniques in the sentiment classification process, it is based on the articles published through ScienceDirect and Springer databases in the interval from 2016 to 2020.It sheds the light on different DL algorithms used, different applications of SA systems. 58 articles studied in ScienceDirect While 26 articles in Springer satisfying the same criteria with the total of 84 articles studied and analyzed in this review. The review concerns with DL techniques, language, domain, and performance results.