Detecting Shill Bidding in Online English Auctions

The popularity of online auctioning has grown meteorically since the late 90s. This attraction is due to their convenience, low cost and ability to reach large (even worldwide) audiences. However, despite the advantages there are many problems inherent in auctioning online. One such problem is shill bidding where spurious bids are introduced into an auction to drive up the final price for the seller, thereby defrauding legitimate bidders. This paper presents an algorithm to detect the presence of shill bidding in online auctions. It observes bidding patterns over a series of auctions, providing each bidder a score indicating the likelihood of their potential involvement shill behaviour. The algorithm has been tested on data obtained from a series of realistic simulated auctions as well as real data acquired from commercial online auctions. Our results show that the algorithm is able to prune the search space required to detect which bidders are likely to be shills. This has significant practical and legal implications for commercial online auctions (such as eBay) where shilling is considered a major threat. While shilling is recognised as a problem, presently there is little or no established means of defence against shills. This paper presents a framework for a feasible solution, which acts as a detection mechanism and a deterrent.