Extracting product features from chinese customer reviews

E-commerce, or business done on the Internet has become more and more popular. Meanwhile, the number of customer reviews for products on the internet grows rapidly. For a popular product, the number of reviews can be in hundreds. As a result, the problem of ¿opinion mining¿ has seen increasing attention over several years. In this paper, we proposed a statistical method to extract product features from Chinese customer reviews. The method is based on distribution of a candidate word in different domains and within the certain domain. It also takes into account the unbalance size of different product reviews. Experimental results show that it achieves better performance than other methods.