An Optimized SVM Model for Detection of Fraudulent Online Credit Card Transactions

Abstract

In order to identify the credit card fraudulent transactions, in this paper we propose an optimized SVM model for detection of fraudulent online credit card model. The model use non-liner SVM and RBF for the sparse transaction data, and use grid algorithm to determine the optional combination of parameters. There dose not exist local minima problem, the curse of dimensionality, small sample size problem and nonlinear problem. In the experiment of the simulation environment, the proposed model performance better than other models which verifies its feasibility.