A combined approach based on fuzzy AHP and fuzzy inference system to rankreviewers in online communities

Abstract

Online product review communities allow users to share their ideas and opinions about various products andservices. Although online reviews as user-generated content can be considered as an invaluable source of information forboth consumers and rms, these reviews tend to be of very different quality. To tackle the problem of low quality reviews,we address reviewer credibility and propose an innovative framework. The framework comprises ve critical phases forranking reviewers in terms of credibility using a fuzzy analytic hierarchy process (AHP) and fuzzy inference system.To determine the weights of the features, a fuzzy AHP method was applied. In addition, according to the proposedframework, to compute a realistic credibility score based on trustworthiness and expertise, a cognitive approach wasfollowed and a fuzzy inference system was designed. To illustrate an application of the proposed method, we conductedan experimental study using real data gathered from Epinions.