Project Title: Credit Card Fraud Detection
Description: Implemented a credit card fraud detection system using machine learning techniques on a dataset containing transactional information. Utilized various classifiers including Logistic Regression, Random Forest, SVM, Gradient Boosting, and Decision Tree to accurately classify transactions as fraudulent or legitimate based on features such as transaction amount, time, and type. Conducted extensive data exploration, preprocessing, and model evaluation to ensure robust performance. Achieved high accuracy and precision, effectively minimizing false negatives and optimizing fraud detection capabilities.
Lessons Learnt: This project emphasized the importance of rigorous data preprocessing, feature selection, and model evaluation in developing effective fraud detection systems. Additionally, it highlighted the significance of leveraging ensemble methods and appropriate scaling techniques to enhance model performance and achieve reliable predictions.
Data Source - https://www.kaggle.com/datasets/willianoliveiragibin/high-fidelity-fraudulent-activity-2023/data
- Accuracy: 0.96 - Precision: 0.98 - Recall: 0.94 - F1 Score: 0.96 - ROC AUC: 0.99
- Accuracy: 1.00 - Precision: 1.00 - Recall: 1.00 - F1 Score: 1.00 - ROC AUC: 1.00
- Accuracy: 0.97 - Precision: 0.99 - Recall: 0.96 - F1 Score: 0.97 - ROC AUC: 1.00
- Accuracy: 0.98 - Precision: 0.99 - Recall: 0.97 - F1 Score: 0.98 - ROC AUC: 1.00
- Accuracy: 1.00 - Precision: 1.00 - Recall: 1.00 - F1 Score: 1.00 - ROC AUC: 1.00
CPU times: total: 2h 44min 52s
Wall time: 3h 42min 13s
Data distribution -