Prevention of credit card fraud detection based on HSVM

Specific crime in the banking system is credit card fraud. Credit card usage has been increased due to the rapid growth of E-commerce techniques. Credit card fraud also increased at the same time. Prevention is better than detection. So the existing system prevented the credit card fraud by identifying fraud in the application of the Credit card. Due to the limitation of the existing system, this paper proposed new algorithm along with the existing algorithm. Scalability issues, extreme imbalanced class and time constraints are the limitation of existing systems. Those limitations are overcome by hybrid support vector machine (HSVM) along with communal and spike detection for credit card application fraud detection. HSVM is the most used method for the pattern recognition and classification.