Comparative Sentiment Analysis on Product Reviews using Deep Learning Techniques

Abstract

Micro blogging websites are nothing but social media site to which user makes short and frequent posts. Twitter is one of the famous micro blogging services where user can read and post messages which are 148 characters in length. Twitter messages are also called as Tweets. We will use these tweets as raw data. We will use a method that automatically extracts tweets into positive, negative or neutral sentiments. By using the sentiment analysis the customer can know the feedback about the product or services before making a purchase. The company can use sentiment analysis to know the opinion of customers about their products, so that they can analyze customer satisfaction and according to that they can improve their product. Now-a-days social networking sites are at the boom, so large amount of data is generated. Millions of people are sharing their views daily on micro blogging sites, since it contains short and simple expressions. In this paper, we will discuss about a paradigm to extract the sentiment from a famous micro blogging service, Twitter, where users post their opinions for everything. We can use the deep learning algorithm to classify the twitters which includes Convolutional Neural Networks. The experimental result is presented to illustrate the use and effectiveness of the proposed system.