Anticipatory Measure for Auction Fraud Detection in Online
Narasamma S1, Suma Latha. K2, Suma Latha. M3
1Narasamma. S, Computer Science and Engineering, Jawaharlal Nehru Technological University, Hyderabad, India.
2Sumalatha. K, Computer Science and Engineering, Jawaharlal Nehru Technological University, Hyderabad, India.
3Sumalatha. M, Computer Science and Engineering, Jawaharlal Nehru Technological University, Hyderabad, India.
Manuscript received on July 20, 2013. | Revised Manuscript received on August 16, 2013. | Manuscript published on August 30, 2013. | PP: 320-324 | Volume-2, Issue-6, August 2013. | Retrieval Number: F2078082613/2013©BEIESP
Open Access | Ethics and Policies | Cite
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: This paper introduces and presents the Online Modeling of Proactive Moderation System for Auction Fraud Detection by Using online feature selection, stochastic search variable selection (SSVS),coefficient bounds from human knowledge and multiple instance learning. An important usability goal of proactive moderation systems is by applying expert knowledge, such as bounding the rule based feature weights to be positive and multiple instance learning, can significantly improve the performance in terms of detecting more frauds and reducing customer complaints given the same workload from human experts.
Keywords: Online Auction, Fraud Detection, Online Modeling, Online Feature Selection, Multiple Instance Learning.