Fast and Effective Credit Card Fraud Detection in Imbalanced Data using Parallel Hybrid PSO

Credit card fraud detection has been one of the major necessities of the current e-commerce based world. The ease of use provided by e-commerce transactions is hindered by the threat caused by fraudsters. Several models have been proposed for identifying fraudulent transaction in a credit card system. However, the threats still do tend to exist. This paper discusses and analyzes the major reasons for credit card models to fail and also proposes a metaheuristic based detection technique to overcome the problems and provide effective detections. This work proposes a parallel PSO based hybrid technique that incorporates Simulated Annealing to provide time effective predictions that are also more accurate compared to conventional techniques.