Back propagation networks for credit card fraud prediction using stratified personalized data

Authors:
Rong-Chang Chen;Chih-Yi Lai
Affiliations:
Department of Logistics Engineering and Management, National Taichung Institute of Technology, Taichung, Taiwan, ROC;Department of Logistics Engineering and Management, National Taichung Institute of Technology, Taichung, Taiwan, ROC
Venue:
ISP'06 Proceedings of the 5th WSEAS International Conference on Information Security and Privacy
Year:
2006

Citing 8
Cited 0

Advances in predictive models for data mining

Pattern Recognition Letters - Special issue on machine learning and data mining in pattern recognition
Distributed Data Mining in Credit Card Fraud Detection

IEEE Intelligent Systems
Neural Data Mining for Credit Card Fraud Detection

ICTAI '99 Proceedings of the 11th IEEE International Conference on Tools with Artificial Intelligence
Minority report in fraud detection: classification of skewed data

ACM SIGKDD Explorations Newsletter - Special issue on learning from imbalanced datasets
Neural network techniques for financial performance prediction: integrating fundamental and technical analysis

Decision Support Systems - Special issue: Data mining for financial decision making
Data Mining: Concepts and Techniques

Data Mining: Concepts and Techniques
Impacts of internet stock news on stock markets based on neural networks

ISNN'05 Proceedings of the Second international conference on Advances in neural networks - Volume Part II
Novel questionnaire-responded transaction approach with SVM for credit card fraud detection

ISNN'05 Proceedings of the Second international conference on Advances in neural networks - Volume Part II

Quantified Score

Hi-index	0.00

Visualization

Abstract

A personalized approach (PA) has been presented recently to prevent fraud in using credit cards. This new approach proposes to predict a user's new transactions by his/her personalized model instead of using multiple-user approaches (MUA) which are based on transaction data of many other users. This approach has shown its potential to deal with the credit card fraud problem. The purpose of this paper is to investigate the performance of back propagation networks (BPN) on predicting credit card fraud using PA. The trained and tested data are stratified, i.e., each class has representative data. To facilitate the decision of the network architecture, Bubble charts are employed. Results from this study show that with stratified data, BPN can obtain good prediction performance. In addition, Bubble chart is a convenient tool to help decide the architecture of the network.