MINING E-COMMERCE FEEDBACK COMMENTS FOR COMPUTING MULTI-DIMENSIONAL TRUST

We propose Comment-based Multi-dimensional trust (CommTrust), a fine-grained multi-dimensional trust evaluation model by mining e-commerce feedback comments. With CommTrust, comprehensive trust profiles are computed for sellers, including dimension reputation scores and weights, as well as overall trust scores by aggregating dimension reputation scores. To the best of our knowledge, CommTrust is the first piece of work that computes fine-grained multi dimension trust profiles automatically by mining feedback comments. In later discussions, we use the terms reputation score and trust score interchangeably. In CommTrust, we propose an approach that combines dependency relation analysis, a tool recently developed in natural language processing (NLP) and lexicon-based opinion mining techniques to extract aspect opinion expressions from feedback comments and identify their opinion orientations. We further propose an algorithm based on dependency relation analysis and Latent Dirichlet Allocation (LDA) topic modelling technique to cluster aspect expressions into dimensions and compute aggregated dimension ratings and weights. We call our algorithm Lexical-LDA. Unlike conventional topic modelling formulation of unigram representations for textual documents our clustering is performed on the dependency relation representations of aspect opinion expressions. As a result we make use of the structures on aspect and opinion terms, as well as negation defined by dependency relations to achieve more effective clustering. To specifically address the positive bias in overall ratings, our dimension weights are computed directly by aggregating aspect opinion expressions rather than regression from overall ratings.