Internet Auction Fraud Detection Using Social Network Analysis and Classification Tree Approaches
International Journal of Electronic Commerce
Generating realistic online auction data
AI'12 Proceedings of the 25th Australasian joint conference on Advances in Artificial Intelligence
Detecting online auction shilling frauds using supervised learning
Expert Systems with Applications: An International Journal
Hi-index | 0.00 |
The online auction is a well-known business model that shows how business can be changed with the aid of new technologies. On the other hand, although the online auction allows buyers to find a wider variety of items and helps sellers to extend to literally millions of buyers, it is also accompanying by a great deal of online auction fraud through the information asymmetry and anonymity problems. As serious online fraud, we deal with the online credit card phantom transaction, which is a fake transaction by the collusion of the seller and buyer using credit card. Basically it is illegal online loan sharking which incurs various social and economic costs: tax evasion, the development of a black-market for the loan service, destruction of debtor's (buyer) financial condition due to the excessively high interest rate. In this paper, we investigated the factors necessary to detect phantom transaction. Based on the studies that have explored the behaviors of buyers and sellers in online auctions, we used the followings as the independent variables: starting bid, auction length, bid count, bid increment, and seller credit. Through empirical bidding data analysis, our logistic regression model suggests the use of "starting bid', "auction length' and "seller credit' as important factors for detection of phantom transaction.