Real-Time Collusive Shill Bidding Detection in Online Auctions

Shill bidding is where a seller introduces fake bids into an auction to artificially inflate an item's final price, thereby cheating legitimate bidders. Shill bidding detection becomes more difficult when a seller involves multiple collaborating shill bidders. Colluding shill bidders can distribute the work evenly among each other to collectively reduce their chances of being detected. Previous detection methods wait until an auction ends before determining who the shill bidders are. However, if colluding shill bidders are not detected during the auction, an honest bidder can potentially be cheated by the end of the auction. This paper presents a real-time collusive shill bidding detection algorithm for identifying colluding shill bidders while an auction is running. Experimental results on auction data show that the algorithm can potentially highlight colluding shill bidders in real-time.