Fuzzy logic based sentiment analysis of product review documents

Sentiment Analysis is the process of determining subjectivity, polarity and polarity strength of a piece of text. Survey shows that 81% of Internet users have done on-line research on a product at least once. Manual analysis of this reviews is difficult. To mine the overall sentiment or opinion polarity, sentiment analysis can be used to mine the overall sentiment or opinion polarity of the review. The objective of this paper is to extract features from the product reviews and classify reviews into positive, negative and neutral. Feature based sentiment classification approach is used to extract sentiments. It involves preprocessing to remove noise, extraction of features and corresponding descriptors and tagging their polarity. The proposed technique extends the feature based classification approach to incorporate the effect of various linguistic hedges. This approach uses fuzzy functions to emulate the effect of modifiers, concentrators and dilators. The system was evaluated with SFU corpus and the results suggest that sentiment analysis using fuzzy logic performs remarkably well.