Analysis of Credit Card Fraudulent Transactions Using Machine Learning and Artificial Intelligence
Sriprada Ramesh, T M Simna, Mohana · 2024
The reliance on credit cards for everyday purchases has become ubiquitous in modern society. This poses security threats as cybercriminals are constantly adopting new techniques to commit fraudulent activities. The need for financial organizations to protect their customer's data is at most important. Machine learning (ML) algorithms provide robust fraud detection capabilities and hence can be applied in in developing advanced systems that continuously monitor transactions in real-time, swiftly identifying suspicious patterns indicative of fraudulent activity. Proposed work focuses on comparing for ML models such as Random Forest, Logistic Regression, SVM, and XGBOOST for detecting credit card frauds. The objective is to maximize the detection of fraud transactions while reducing false positives. The f1 score of the models are compared to identify the most efficient algorithm. XGBOOST and Random Forest proved to perform the best in credit card fraud detection with a fl-score of 0.81.