Efficient Contextual Transaction Trust Computation in E-commerce Environments

Haibin Zhang, Yan Wang, Xiuzhen Zhang

Abstract: In e-commerce environments, trust is a dominating factor in seller selection. Most existing trust evaluation studies compute a single value to reflect the "general' or "global" trust level of a seller provider without any contextual transaction information taken into account. As a result, a buyer may be easily deceived by a malicious seller in a forthcoming transaction. For example, with the notorious "value imbalance problem", a malicious seller can build up a high trust level by selling cheap products and then starts to deceive buyers in selling expensive products. To detect this problem and avoid massive monetary losses of buyers, trust evaluation should be associated with both past transactions and the new one, and take transaction context into account. In particular, the computed trust result should outline the seller's reputation profile indicating the trust level in a specific product or a product category, a price range, a time period or any necessary combination of them. However, this need requires complex computation and thus new data structures and efficient algorithms. In this paper, we design a new data structure to support the CTT computation in e-commerce environments. In addition, based on the new data structures, we further propose an approach for promptly responding to a buyer's CTT query. The conducted experiments illustrate that our proposed structure and approach can yield much shorter computation time than the existing approaches.