Enhancing E-Commerce Security: A Machine Learning Framework for Fraud Detection

A business paradigm known as electronic commerce, or e-commerce, enables businesses and individuals to purchase and sell anything over the internet. Since frauds been for a long and erratic and lack consistency, it might be challenging to spot them. The drastic changes are exploited by con artists. Huge looses are been occurred due a result of their devious manipulation of security checks. An increase in internet customers is contributing to a spike in transactions. We note that deception is becoming more common in online transactions as well. The use of device learning in online commerce will increase to prevent deception. This research looks into machine learning methods for detecting an fraud for various e-commerce, with a particular emphasis on decision trees, RF, and Neural Net. Based on a dataset of 149,156 records—13,962 of which are flagged as fraudulent—obtained from Kaggle, the study applies a quantitative methodology to assess categorization results.