As a new type of word-of-mouth information, online consumer reviews possess critical information regarding consumer‘s concerns and their experience with the product or service. Such information is considered essential to firms‘ business intelligence which can be utilized for the purpose of production recommendation, personalization, and better customer understanding. This paper considers the problem of online reviews sentiment mining based on the theory of consumer psychology and behavior. Given the fuzzy attribute nature of the online reviews, we have established fuzzy group bases of consumer psychology. Four fuzzy bases, including features, sense, mood and evaluation, are established. The consumer attitude elements are reflected by natural language reviews. A fuzzy sentiment computing algorithm of online reviews for consumer sentiment is developed, and a fuzzy rule base is also presented based on consumer decision-making process. Finally it shows by means of an experiment that the proposed approach is very well suited as an analysis tool for the online reviews sentiment mining problem.
Zhao, Narisa; Li, Yuan; and Wang, Jianjun, "Using Fuzzy Sentiment Computing and Inference Method to Study Consumer Online Reviews" (2010). ICEB 2010 Proceedings. 44.