Monitoring Suspicious Discussions and Evaluation of Products on Online Forums Using Text Mining

Abstract

Along with the rapid growth of the Internet teclmology, interaction among people is increased and it reduced the distance among them. Thousands of forums are available that provide huge virtual space to share and discuss their opinions and express their views about any aspect of life. These forums also provide platform to consumers who want to express their opinions about product that can depict the behaviour of people toward the product. Analysis of their opinion can become an important instrwnent for performance evaluation of product quality and market research. Fwthermore, opinions towards competing products are helpful to measme sale volume of both produc~s. Although, there are various benefits of online forum, there are also some drawbacks that are undesirable featmes of forums. Some of these are, use of unethical language for comments, online exchange of suspicious post in the form of text, image or video. Some previous work done on word clustering and evaluation of things on basis of rating or search percentage done by others': Consider a person who returns from an extended vacation and needs to find out quickly what happened in the world during her absence. Reading the entire news collection is a daunting task, and generating specific queries about unknown facts is rather unrealistic. Thus, intelligent assistance from the computer is clearly desirable. Such assistance could take the form of a content sununary of a corpus for a quick review, the temporal evolution of past events of interest, or a listing of automatically detected new events which demonstrate a significant content shift from any previously known events. It would also be useful to have structured guidelines for navigation tlu-ough document clusters. This project is aim to develop web application that would be used to evaluate the product/ Article based on the posting. This evaluation shows how customers/users rate the product. This system will use text-mining teclmiques to analyse online plain text from any selected forum, evaluate the comments of customer, and classify them positive, negative and neutral. Based on these evaluation products will be rated. This application will also identify w1ethical cotnments and use of unethical language and block them before posting to blog. This will help to avoid and reduce illegal activities on internet.