Credit Card Fraud Detection Using Logistic Regression with Hyperparameter Optimization

Credit card fraud detection is a significant challenge in the financial industry. In this research, a logistic regression model is implemented to detect fraudulent credit card transactions, with a focus on hyperparameter optimization and handling class imbalance. <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$U$</tex>sing Grid Search technique to find the best hyperparameter combination and SMOTE oversampling to handle class imbalance. Experimental results show that the optimized model achieves significant performance improvement over the baseline model. The baseline model showed a precision of 57.42<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">%</sup>, recall of 92.15%, F1-Score of 62.37%, accuracy of 99.13%, and ROC-AUC of 97%. After optimization, the model achieved a precision of 89.95%, recall of 86.33%, F1-Score of 88.06%, and accuracy of 99.83%. This improvement shows that the hyperparameter optimization and class imbalance handling techniques used are able to improve the effectiveness and accuracy of credit card fraud detection.