Industry: adaptive fraud detection
暂无分享,去创建一个
This article describes the development of a prototype data mining system for detecting cellular phone (cloning) fraud. In cellular cloning fraud, the identity of a legitimate cellular phone is programmed into another; from the second phone, calls can be made illicitly that are charged to the customer's account. The system for detecting such fraud is based on a framework that uses a sequence of data mining techniques. First, a rule learning program discovers general indicators of fraudulent behavior from a large database of defrauded accounts. Next, the indicators are used to create a set of monitors, which profile customer behavior and measure anomalies. Finally, the outputs of the monitors are assigned weights by a linear threshold unit. Experiments with the system indicate that this automatic approach performs better than hand-crafted methods for detecting fraud. Furthermore, the system can be retrained as necessary to accommodate changing conditions of fraud detection environments.
[1] Kathleen Martin,et al. The Learning Machines. , 1981 .
[2] Tom Fawcett,et al. Activity monitoring: noticing interesting changes in behavior , 1999, KDD '99.
[3] Tom M. Mitchell,et al. MODEL-DIRECTED LEARNING OF PRODUCTION RULES1 , 1978 .
[4] Tom Fawcett,et al. Adaptive Fraud Detection , 1997, Data Mining and Knowledge Discovery.
[5] Ron Kohavi,et al. The Case against Accuracy Estimation for Comparing Induction Algorithms , 1998, ICML.