ABSTRACT

With the popularity of online shopping it is increasingly becoming important for manufacturers and service providers to ask customers to review their product and associated service. Typically the number of customer reviews that a product receives grows rapidly and can be in hundreds or even thousands. This makes it difficult for a potential customer to decide whether to buy the product or not. It is also difficult for the manufacturer of the product to keep track and manage customer opinions. Opinion mining is an emerging field that classifies a user opinion into positive and negative reviews. In this paper it is proposed to develop a methodology using word score based on Singular Value Decomposition by modeling a custom corpus for a given topic in which opinion mining has to be performed. Bayes Net and decision tree induction algorithms are used to classify the opinions.