Statistical Features Identification for Sentiment Analysis Using Machine Learning Techniques

Abstract

Due to increasing fascinating trend of using internet and online social media, user-generated contents are growing exponentially on the Web, containing users' opinion on various products. In this paper, we have proposed a sentiment analysis system which combines rule-based and machine learning approaches to identify feature-opinion pairs and their polarity. The efficiency of the proposed system is established through experimentation over customer reviews on different electronic products.