A bio-inspired credit card fraud detection model based on user behavior analysis suitable for business management in electronic banking

The widened uses of Internet credit cards in e-banking systems are currently prone to credit card fraud. Data imbalance also poses a significant difficulty in the method of fraud detection. The efficiency of the existing fraud detection systems is only in question because it detects fraudulent action after the suspect transaction has been completed. To address these difficulties, this article offers an improved two-level credit card fraud tracking model from imbalanced datasets based on the semantic fusion of k-means and the artificial bee colony (ABC) algorithm to improve identification precision and accelerate the convergence of detection. In the proposed model, ABC works as a kind of neighborhood search associated with a global search to be a second classification level to manage the failure of the k-means classifier to explore the actual clusters as it is sensitive to the initial condition. The proposed model filters the characteristics of the dataset using an integrated rule engine to evaluate whether the operation is real or false, depending on many parameters of client conduct (profile) such as geographical locations, usage frequency, and book balance. Experimental findings show that the suggested model can improve the precision of ranking against the danger of suspect operations and provide higher accuracy relative to traditional techniques.