Product rating generation based on public opinion using sentiment analysis

Abstract

In modern days, people tend to check reviews and opinions on a product before buying. The main goal of our system is according to the public opinion of a product, give a rating to the product on a scale of 0 to 10. Also, we are plotting this rating for better understanding of how every attribute of a product stands against time. For public opinion, we collected data from twitter. The reason behind using twitter tweets is because it is one of the most popular social sites. The data is pre -processed and then filtered for irrelevant characters. The data is then clustered based on different attributes of product. After that using the Naïve Bayes classifier we do a sentiment analysis of the data to calculate the polarity. This polarity is then converted to a scale of 0 to 10 (where 5 is average) and thus the rating of an individual product is obtained. This polarity for each product for every attribute is plotted in a graph where axis-x represents time and axis-y represents polarity.