Description

The proposed CommTrust, a multi-dimensional trust evaluation model, for computing comprehensive trust proles for sellers in e-commerce applications. Different from existing multi-dimensional trust models, we compute dimension trust scores and dimension weights automatically via extracting dimension ratings from feedback comments. Based on the dependency relation parsing technique, we have proposed Lexical-LDA (Lexical Topic Modeling based approach) and DR-mining (Lexical Knowledge based approach) approaches to mine feedback comments for dimension rating proles. Both approaches achieve significantly higher accuracy for extracting dimension ratings from feedback comments than a commonly used opinion mining approach. Based on the opinion that buyers often express thoughts openly in free text feedback annotations, propose CommTrust for trust valuation by mining feedback comments. The propose a multidimensional trust model for subtracting reputation cuts from user feedback observations. This enquiry is the first piece of work on trust valuation by mining pointer comments. The reputation score for a seller is the positive section score, as the percentage of positive rankings out of the total number of encouraging ratings and negative ratings. One possible reason for the lack of negative grades at e-commerce web sites is that users who leave negative advice ratings can attract vengeful negative ratings and thus damage their own standing. Extensive research on eBay and Amazon data determine that CommTrust can excellently address the “all good reputation” issue and rank sellers meritoriously. The score given by model is constant with the artificial for most evaluations, and lower disbelieving reviews’ score.