A feature based approach for sentiment analysis using SVM and coreference resolution

Abstract

Online shopping is one of the most comfortable ways to shop in this new era of technology. People buy online products frequently and post their reviews about the products they have used. The viewpoint of the user will be in the form of tweets or product reviews which they post in an e-commerce site. These reviews will have significant role in deciding how far the products have been placed in peoples mind. These reviews will also help the manufacturers to improve the features of the product as required But it is a very difficult task to manually read the reviews and assign sentiment to them. This problem can be solved by creating an automated system in which we can analyze the reviews posted by the users and extract the users perception about a particular feature. In this work we have developed a procedure for 'feature based sentiment analysis' by using a classifier called Support Vector Machine.