Weighted fuzzy rule based sentiment prediction analysis on tweets

Abstract

As E-Commerce is becoming more popular, the number of product reviews that a product received grows exponentially. In this context, others’ opinions will play a vital role to make a decision to select among multiple options involves valuable resources like money and time, where people usually depends on their peers’ past experiences in the form of reviews. Many companies use opinion mining and sentiment analysis as part of their research. Detecting sentiment in social media like Facebook, Twitter, online blogs and reviews have become an essential task as they have been influencing every business organization. In this paper, we would like to analyze the Fuzzy rule-based systems (FRBSs) with FRBS models, namely, Mamdani and Takagi Sugeno Kang (TSK) using FRBS package in R, and a comparison with other common classification methods. Additionally, the performance of FRBSs method is measured in terms of precision, Recall and F-measure to find accuracy in classification.