Multi-dimensional e-commerce trust evaluation method

In the last decade, e-commerce has been grown rapidly and become a familiar tool of shopping for many people. However, some people still have concerns while making online purchases due to its uncertain attributes. In fact, there are many online consumers have suffered from monetary loose problem due to some reasons which the lack of the trust in e-commerce is one of them. Therefore, there is a great demand for a mechanism that helps to evaluate the trust throughout the online transactions. One of them is the existing mechanism of the trust management which is used in some e-commerce websites (e.g. eBay). Such a mechanism evaluates the trust by computing a trust value of any seller only based on the previous rating of the past transactions. Therefore, the trust value is only able to show the general status of the trust without taking into the account the new transaction. Consequently, there is a great possibility for the frauds to be committed by some of the malicious people. For example, some of them can easily build a good reputation by making many transactions by selling cheap products with good qualities and start to commit frauds by selling more expensive products. This kind of frauds is named by [1] as the value imbalance problem. Therefore, there is a great demand for a trust evaluation mechanism which consider the new transaction as well as the past transactions. In this paper, we propose a new method which considers three dimensions that play important roles in any online transaction to help the buyers to detect the frauds. This method measures the similarity between the new transaction and the past transactions in the products types dimension, the number of the products sold dimension and the transactions amounts dimension.