Experimental Evaluation of Identifying Fraudulent Activities in E-Commerce to Protect Consumers by Using Intelligent Deep Learning Methodology

To make use of and analyze the vast quantities of data produced by online transactions, advanced cyber-infrastructure and information technology approaches are required. In this research, we provide a big data platform that online stores may use to address a number of problems plaguing the e-commerce sector. Businesses and individuals alike are susceptible to fraud, a global issue. Deep Learning and AI have been tremendous assets in the fight against fraud in today's tech-driven world. Electronic wallets and credit cards were among the first digital payment options offered with the rise of online marketplaces. The goal of implementing a fraud detection system is to identify and prevent fraudulent transactions that involve stolen or unverified credit card information. With the use of a deep learning algorithm, this work intends to offer substantial assistance, leading to the creation of a new model known as the Deep Regression Classification Model (DRCM). To assess the efficacy of the suggested model, it is cross-validated with the typical learning method known as Naive Bayes (NB). Electronic commerce platforms may improve their security posture, increase user trust and confidence, and reduce financial losses by implementing AI-powered fraud protection methods. This groundbreaking method represents a sea change in the fight against online fraud and highlights the revolutionary power of AI to protect consumers and companies. A system that employs an effective deep learning technique for precise fraud detection will be conceptualized and developed based on the results of the experimental assessment.