Paper Title
A Fuzzy Approach For Sentiment Analysis Based On Web Customer Reviews
Abstract
Sentiment analysis has become an important research area in today’s competitive business world. Customers now
no longer ask friends or relatives for their suggestions to buy a product; everything is available just at a click on the World
Wide Web. Opinions expressed in the user comments on the have become an important source for analysis e.g., customer
reviews of products, ratings, posts, and blogs. In this paper, we focus on customer reviews of products. Specifically, we
study the sentiments/opinions expressed (positive, negative or neutral) in the customer reviews. Hereby, we propose a new
method i.e. fuzzy ontology tree for giving a more clear and precise analysis of sentiments to give best quality poll to
customers. This approach allows the system to handle opinion words that are context dependent, which cause major
difficulties for existing algorithms. In the reviews, a single sentence may exhibit multiple aspects for opinions. Bootstrapping
algorithm is used to handle this problem. The proposed method doesn’t require labelled data and hence easy to implement.