Selecting appropriate sellers in online auctions through a multi-attribute reputation calculation method

Online auctions have become immensely popular and created massive cash turnover in recent years. The volume of trade on eBay, the largest auction site in the world, reached US$6 billion in 2008. However, for a user intent on purchasing an item from an auction site, selecting an appropriate seller from the numerous choices is not an easy task. Even though most auction sites provide a concise binary reputation management mechanism to model the reputation of a trader through an integer value rating system, such a simple mechanism does not give users enough information about their potential trading partners. It is difficult to infer the right judgment rule correctly from knowledge of summing positive and negative ratings alone. We focus on developing an effective reputation model for online auctions to help users select a suitable seller. To accomplish this, four feature factors strongly related to online auction characteristics are adopted to assess the reputation of a trader. We also propose a multi-attribute reputation management (MARM) support tool to assist users in choosing sellers when using auction sites. In this research, actual transaction data collected from eBay were used to demonstrate the effectiveness of our method. Our results show that MARM is able to select more suitable sellers than other methods.