Real-time Shill Bidding Fraud Detection Empowered with Fussed Machine Learning
- Authors
- Abidi, W.U.H.; Daoud, M.S.; Ihnaini, B.; Khan, M.A.; Alyas, T.; Fatima, A.; Ahmad, M.
- Issue Date
- Jul-2021
- Publisher
- Institute of Electrical and Electronics Engineers Inc.
- Keywords
- Artificial neural networks; Data models; deep learning model; e-Auction fraud; Hidden Markov models; Labeling; online fraud detection; Real-time systems; Shill Bidding; Support vector machines; Training
- Citation
- IEEE Access, v.9, pp.113612 - 113621
- Journal Title
- IEEE Access
- Volume
- 9
- Start Page
- 113612
- End Page
- 113621
- URI
- https://scholarworks.bwise.kr/gachon/handle/2020.sw.gachon/81933
- DOI
- 10.1109/ACCESS.2021.3098628
- ISSN
- 2169-3536
- Abstract
- Shill Bidding (SB) occurs when the fake bidders are introduced by the seller’s side to increase the final price. SB is a crime committed during the e-Auction, and it is pretty difficult to detect because of its normal bidding behaviour. The bidder gets a lot of loss because he pays extra money, and the sellers benefit from shill bidding, so this article proposed a fusion base model. This proposed model is split into two parts training and validation, into 70 and 30 per cent. This model is divided into three sub-models, first two models are Support vector machine (SVM) and Artificial neural network (ANN) that are trained parallel on the same dataset and predict the bidding fraud. The prediction of these models becomes the input of the fuzzy-based fussed module, and fuzzy decide the actual output based on SVM and ANN predictions. On every bid, it predicts whether the fraud is committed or not. If the bidding behaviour is normal, then continue the bidding; otherwise, cancel the bid and block the user. The prediction accuracy of the proposed fussed machine learning approach is 99.63%. Simulation results have shown that the proposed fussed machine learning approach gives more attractive results than state-of-the-art published methods. CCBY
- Files in This Item
- There are no files associated with this item.
- Appears in
Collections - ETC > 1. Journal Articles
Items in ScholarWorks are protected by copyright, with all rights reserved, unless otherwise indicated.