Credit Card Fraud Detection System

Credit card fraud poses a significant threat to financial security, leading to substantial monetary losses and risks for both consumers and financial institutions. Detecting fraudulent transactions accurately is challenging due to the high class imbalance in credit card transaction datasets. This study employs machine learning techniques to detect fraudulent transactions using the publicly available creditcard.csv dataset. Data preprocessing steps such as feature scaling, handling imbalanced data, and exploratory data analysis (EDA) are applied to enhance model performance. The study primarily utilizes Logistic Regression as a baseline model, comparing its performance with other machine learning classifiers such as Random Forest, Decision Tree, and Support Vector Machine (SVM). The evaluation metrics include accuracy, precision, recall, F1-score, and AUC-ROC curve analysis to ensure effective fraud detection. The results indicate that balancing the dataset and selecting optimal models can significantly improve fraud detection performance. This research contributes to advancing machine learning-based fraud detection systems and enhancing financial transaction security.