Fraud Detection in Credit Card Transactions Using Neural Networks and Firefly Optimization

Abstract

       In the ever-evolving world of e-commerce and digital transactions, credit card fraud has emerged as a significant challenge. Traditional fraud detection techniques are often inefficient due to their inability to handle vast amounts of data in real-time. This report explores the development of a hybrid fraud detection system using neural networks (NN) combined with firefly optimization algorithms. The NN provides a robust mechanism to detect fraudulent patterns in data, while firefly optimization enhances the detection efficiency by optimizing the weights and structure of the NN. This hybrid approach improves the accuracy, detection speed, and generalization capability in identifying fraudulent credit card transactions.