An Enhanced Credit Card Transaction for Outlier Fraud Detection System Using Artificial Neural Network (ANNs)

Credit card fraud (CCF) is a recurrent problem all over the world. Some countries, despite having a low or average use of credit card, have a high percentage of credit card fraud recorded. The use of credit card that is (cashless) economy is important to any growing nation. Outlier detection aims at finding patterns in data that do not conform to expected behavior in the credit card transaction; it also has extensive use in a wide variety of applications such as military surveillance for enemy activities, intrusion detection in cyber security, fraud detection for credit cards, insurance or health care and fault detection in safety critical systems. In this work, we developed an enhanced credit card transaction for outlier fraud detection system using an artificial neural network algorithm called cortical learning algorithm and customer self-strip detection (OTP, Account Number and BVN) for detecting fraud from the user (the customers) perspective. In the admin column, it converts the highly populated data to a sparse polar representation. Support Vector Machine was used to train active data on the customers' column which is our area of concentration. Structured System Analysis and Design Methodology (SSADM) and PHP programming language are used in implementing .MySQL and Simplified unified database are used for NUBAN for every report made from the customer. The parameters for our result performance achieved an overall performance rate of 95% when compared with the most recent Outlier Fraud Detection System for Flight Reservation Booking. The parameters for the comparison included Time Complexity (TC), Life-Cycle Assessment (LCA), Benchmarking (B), Multi-Criteria Decision Making (MCDM), Risk Assessment (RA), Cost Benefit Analysis (CBA) and Speed (S) presented as TC, LCA, B, MCDM, RA, CBA, S = 21, 15, 7, 22, 8, 10, 12 respectively.