Feature-Opinion Pairs Discovery for Chinese Review
Xiu Li, Huimin Wang, Lulu Xie · 2016
Feature-based opinion mining for product review is the field of study that analyzes user's attitude towards product attributes, which has been witnessed a booming interest in the last one and half decades, due to its importance to business and society as a whole. This paper proposed a POS patterns matching method to identify feature words, opinion bearing words, as well as negative words based on sentence compression. Chinese review text of two type of product: hedonic and utilitarian were used to find the patterns and tested the methods. The experiment results showed that the proposed method perform well.