Shill bidding is where spurious bids are introduced into an auction to drive up the final price for the seller, thereby defrauding legitimate bidders. While shilling is recognized as a problem, presently there is little or no established means of defense against shills. This chapter 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 his/her potential involvement in shill behavior. The algorithm has been tested on data obtained from a series of realistic simulated auctions, and commercial online auctions. 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 where shilling is considered a major threat. This chapter presents a framework for a feasible solution, which acts as a detection mechanism and a deterrent.
This section explores some related work that has been conducted into shilling. At present there is limited coverage on how to detect shill bidding.
eBay has been involved in many legal disputes where bidders/sellers have been accused of shilling (see Schwartz and Dobrzynski (2002)). eBay has clear rules regarding shill bidding behavior in their auctions. Their policy clearly outlines undesirable bidder behavior and the penalties for shill bidding. The regular process for a bidder who suspects that they have been shilled is to contact eBay, who then investigates the incident. eBay does not state exactly what factors they use to determine whether shilling has occurred, nor how to detect which bidders are shills.
Shah et al (2003) use data mining techniques to produce evidence of shilling. Their work used data from approximately 12,000 commercial auctions looking for associations between bidders and sellers. Bidders (or groups of bidders) that participated frequently in auctions held by particular sellers were deemed suspect. However, the authors’ state that their analysis is very limited in that it only looks for simple associations. They suggest that a much more thorough analysis must be performed using complex associations which consider a wider range of shill behavior.
There are also companies who offer data mining techniques to detect fraud in online auctions. However, like eBay, these companies have not made their techniques public.
Wang et al (2002) discuss an approach that attempts to deter shilling in the first place. The Auctioneer is allowed to use fees to make shilling unprofitable for the seller. In auctions using a reserve price, a seller is charged an increasing fee based on how far the winning price is from the reserve price.
Aspects regarding the economic theory of shilling have been examined by Kauffman and Wood (2003), and Barbaro and Bracht (2005).Top
Online English Auctions
This section describes the operation of online English auctions, which is the auction model used in this paper. Formally an English auction can be defined as an ascending-price, open-bid auction. Each bid submitted must be higher than the current highest bid. The minimal amount required to outbid others is usually a percentage of the current highest bid. The value of the current highest bid is available to all parties, along with the auction timing. The winner is the bidder with the highest bid when the auction terminates.
Key Terms in this Chapter
E-Commerce: A system used to conduct business transactions of buying and selling goods and services over a computer network.
Shill Bidder: A bidder who engages in shill bidding behavior.
Shill Score: A rating between zero and ten that indicates the likelihood that a bidder has engaged in typical shill behavior based on his/her past actions in a series of auctions.
Shill Bidding: Using spurious bids on the seller’s behalf to artificially inflate an auction’s price.
Proxy Bidding: A system offered by Online Auctioneers that allows the bidder to enter the maximum price they are willing to pay, and will incrementally increase his/her current bid when out bid, up to the maximum price specified.
Aggressive Shilling: A shill bidder that consistently outbids any legitimate bid to force the price up as much as possible.
Software Bidding Agents: A program that bids on a human bidder’s behalf.
Benign Shilling: A shill bidder that only uses an initial bid at the auction’s beginning in an attempt to try and stimulate bidding.
Online Auctions: An auction conducted over the Internet.
Complete Chapter List
Manish Gupta, Raj Sharman
C. Warren Axelrod
Ahmed Awad E. Ahmed
Arunabha Mukhopadhyay, Samir Chatterjee, Debashis Saha, Ambuj Mahanti, Samir K. Sadhukhan
Zhixiong Zhang, Xinwen Zhang, Ravi Sandhu
Madhusudhanan Chandrasekaran, Shambhu Upadhyaya
Ghita Kouadri Mostefaoui, Patrick Brézillon
Douglas P. Twitchell
James W. Ragucci, Stefan A. Robila
Nick Pullman, Kevin Streff
E. Kritzinger, S.H von Solms
Donald Murphy, Manish Gupta, H.R. Rao
Sérgio Tenreiro de Magalhães, Kenneth Revett, Henrique M.D. Santos, Leonel Duarte dos Santos, André Oliveira, César Ariza
Carsten Röcker, Carsten Magerkurth, Steve Hinske
Yuko Murayama, Carl Hauser, Natsuko Hikage, Basabi Chakraborty