Building High-Quality Auction Fraud Dataset

Abstract

Given the magnitude of online auction transactions, it is difficult to safeguard consumers from dishonest sellers, such as shill bidders. To date, the application of Machine Learning Techniques (MLTs) to auction fraud has been limited, unlike their applications for combatting other types of fraud. Shill Bidding (SB) is a severe auction fraud, which is driven by modern-day technologies and clever scammers. The difficulty of identifying the behavior of sophisticated fraudsters and the unavailability of training datasets hinder the research on SB detection. In this study, we developed a high-quality SB dataset. To do so, first, we crawled and preprocessed a large number of commercial auctions and bidders' history as well. We thoroughly preprocessed both datasets to make them usable for the computation of the SB metrics. Nevertheless, this operation requires a deep understanding of the behavior of auctions and bidders. Second, we introduced two new SB patterns and implemented other existing SB patterns. Finally, we removed outliers to improve the quality of training SB data. •Bidders attempt to detect SB by themselves by tracking many of the competitor's behavior and communicating their suspicions to eBay. Very recently, the bidders' IDs and history are no longer available on eBay. We believe this new policy about blocking the bidding history is to not be able to discover SB activities. •Buyers are the most affected by SB since they pay much more for the items. The price is driven up by disingenuous bidders with no intention of ever winning the bid. For instance, CNB News disclosed that a bidder paid $1, 825 for a nearly complete set of 1959 Topps baseball cards on eBay (nbcnews.com nd). However, two undercover detectives determined that the purchaser ended up paying an extra $531 for the cards due to SB

Cost-Sensitive Semi-supervised Classification for Fraud Applications

“…We developed a reliable SB dataset using a large collection of commercial auctions of eBay and their bidder history too [7] (see Table 1). We rigorously preprocessed the two crawled datasets, auctions and bidders.…”

“…Then, we implemented a collection of nine SB strategies exposed in Table 1. For more details about the fraud patterns and their measurement algorithms, consult the article [7].…”