Please use this identifier to cite or link to this item: http://hdl.handle.net/10125/50370

A Real-Time Detection Algorithm for Identifying Shill Bidders in Multiple Online Auctions

File SizeFormat 
paper0483.pdf1.42 MBAdobe PDFView/Open

Item Summary

Title: A Real-Time Detection Algorithm for Identifying Shill Bidders in Multiple Online Auctions
Authors: Majadi, Nazia
Trevathan, Jarrod
Keywords: Social Shopping: The Good, the Bad and the Ugly
Auction fraud, Bidding behavior, Online auction, Live Shill Score, Shill bidding.
Issue Date: 03 Jan 2018
Abstract: Online auctions are highly susceptible to fraud. Shill bidding is where a seller introduces fake bids into an auction to drive up the final price. If the shill bidders are not detected in run-time, innocent bidders will have already been cheated by the time the auction ends. Therefore, it is necessary to detect shill bidders in real-time and take appropriate actions according to the fraud activities. This paper presents a real-time shill bidding detection algorithm to identify the presence of shill bidding in multiple online auctions. The algorithm provides each bidder a Live Shill Score (LSS) indicating the likelihood of their potential involvement in price inflating behavior. The LSS is calculated based on the bidding patterns over a live auction and past bidding history. We have tested our algorithm on data obtained from a series of realistic simulated auctions and also commercial online auctions. Experimental results show that the real-time detection algorithm is able to prune the search space required to detect which bidders are likely to be potential shill bidders.
Pages/Duration: 10 pages
URI/DOI: http://hdl.handle.net/10125/50370
ISBN: 978-0-9981331-1-9
DOI: 10.24251/HICSS.2018.482
Rights: Attribution-NonCommercial-NoDerivatives 4.0 International
Appears in Collections:Social Shopping: The Good, the Bad and the Ugly


Please contact sspace@hawaii.edu if you need this content in an ADA compliant alternative format.

Items in ScholarSpace are protected by copyright, with all rights reserved, unless otherwise indicated.