Credit card fraud detection using Machine Learning Algorithms

Credit card fraud has emerged as a critical issue in the modern financial ecosystem due to the exponential rise of online transactions. This paper proposes a machine learning-based framework for fraud detection, leveraging Logistic Regression, Random Forest, and Gradient Boosting algorithms on the Kaggle credit card fraud dataset. Data preprocessing techniques such as feature scaling, normalization, and Synthetic Minority Oversampling Technique (SMOTE) are applied to handle class imbalance. The models are evaluated on metrics including Accuracy, Precision, Recall, F1-Score, and ROC-AUC. Experimental results indicate that ensemble methods, particularly Gradient Boosting, outperform traditional models in fraud detection. This research highlights the potential of ML-based fraud detection systems to enhance security, reduce financial losses, and improve customer trust.