An opinion search system for consumer products

An opinion search system for consumer products is proposed, which utilizes data mining and information retrieval technology and Experimental results on a real-world data set show the system is feasible and effective.

Abstract

With the rapid progress of e-commerce, many people like purchasing product on the e-commerce website, and giving their personal reviews to the product they purchased, so the number of customer reviews grows rapidly. Generally, a potential customer will browse product reviews before they purchase the product. However, retrieving opinions relevant to customer’s desire still remains challenging. To provide efficient opinion information for customers, we propose an opinion search system for consumer products, which utilizes data mining and information retrieval technology. A ranking mechanism taking temporal dimension into account and a method for results visualization are developed in the system. Experimental results on a real-world data set show the system is feasible and effective.