Refining Cyber Fraud Tactics in the Retail Sector Applying Defense Mechanisms for Digital Trade Platforms

Abstract

The proposed work integrates deep learning and blockchain technologies to investigate a hybrid strategy for combating cybercrime in the retail industry. Long Short-Term Memory Networks combined with Convolutional Neural Networks have been used efficiently and effectively for anomaly detection, with a high success rate for fraudulent transactions. The Proof-of-Authority consensus algorithm is used in the blockchain measure, which ensures immediate validity with a transaction throughput of 1,200 transactions per second and validation times averaging 0.9 seconds. It offers a scalable and effective solution for securing digital trade platforms in retail and shows a reduction in fraud over six months.