Sentiment Analysis of Customer Reviews based on Hidden Markov Model

Presently there are various websites like Amazon.com, eBay, FlipKart, Snapdeal etc. which have large number of products available online. The seller or the manufacturer often ask their customers to share their opinions and hands-on experiences on the products they have purchased. Unfortunately, it is very difficult to go through all customer's reviews and to decide whether the overall performance of the product is satisfactory or not. This paper mainly focuses on the problem of sentiment analysis of customer's online reviews about the product. In this paper we aim to train our system to analyse whether the comment given by the customer is positive or negative. The work is divided in two phases: In first phase, we propose a training process of a Stochastic model namely Hidden Markov Model and secondly we test and reveal the individual comment for analyzing consumer opinions about the products. Our results indicates that the trained system is very promising in performing its tasks and we have achieved maximum possible Precision and Accuracy.