FEASA: AN APPROACH TO RESOLVE SENTIMENT AND GENERATE FEATURE EXTRACTION MATRIX

Abstract

Sentiment analysis is the study of classifying human’s sentiments, evaluations, attitudes, opinions about some topic, product, expressed in form of text or speech. As e-commerce is becoming more and more popular, the number of customer reviews that a product receives grows rapidly. For a popular product, the number of reviews can be in hundreds or even thousands. This makes it difficult for a potential customer to read them to make an informed decision on whether to purchase the product. In order to improve the customer satisfaction, many ecommerce sites provides the provision to write reviews about products. Instead of manually reading and evaluating numerous reviews, an automated procedure can be helpful and can be easy to get the overall polarity of the reviews for the product. The goal of the research is to present an approach with the target of deriving qualitative sentiment analysis which will be helpful for recommendation. This paper introduces the FEASA (Feature extraction and sentiment analysis), an approach to resolve sentiment and generate feature extraction matrix for each product. FEM will help in determining the features, being commented upon and also help in detecting the popular features among the reviews. The paper aims at mining sentiment for each product review and summarizes overall sentiment score associated with each product review. The efficient ranking of the features will be deduced, based on the opinions.