A Hybrid Approach of Lexicon-based and Corpus-based Techniques for Arabic Book Aspect and Review Polarity Detection
Raja Masadeh · International Journal of Advanced Trends in Computer Science and Engineering · 2020
The problem of Aspect Based Sentiment Analysis (ABSA) has been studied very well in English language, and there are many approaches have been suggested for ABSA.However, there are other languages did not find enough attention in this field.One of these languages: Arabic language that covers a large area of the world.ABSA contains several tasks.This paper addresses for both aspect polarity and review polarity detection.A hybrid approach proposed to treat polarity detection problem, which consists of corpus-based approach (review book corpus) and lexical-based approach (automatic lexicon) used for this work.The work starts by annotating dataset and cleaning it from a noisy data, after that automatically building the lexicon with sensible words with their polarity that hints to extract aspects with their polarity.Experimental results showed that the proposed system got an acceptable result.