Credit Card Fraud Detection using Machine Learning Models and Increasing Explainability using Explainable AI Methods

Intentional deception and exploitation through credit card fraud remain prevalent in the banking industry, as fraudsters continue to obtain credit card details without authorization. Recent years have seen a significant increase in the number of credit card accounts and expenditure, accompanied by a noticeable rise in fraud instances. For fraud detection, the financial sector has employed various machine learning (ML) models and artificial intelligence (AI) techniques. This research follows a systematic methodology to identify and elucidate instances of credit card fraud. A Credit Card Fraud dataset was used. Machine learning models, including Explainable Boosting Machine (EBM), Decision Tree (DT), Random Forest (RF), XGBoost (XGB) and an Artificial Neural Network (ANN) were instantiated. Explainable AI models such as SHAP, LIME, and EBM were applied to the outputs of these ML models to enhance the explainability. The results indicated that the EBM and Decision Tree models achieved the highest accuracy at 96.35%.