Project Overview: This project aims to build an efficient fraud detection model to identify fraudulent transactions while minimizing false positives. It uses Machine Learning techniques, feature engineering, and class imbalance handling (SMOTE).
Objectives
- Develop a classification model for fraud detection
- Handle class imbalance using oversampling (SMOTE)
- Engineer meaningful features like transaction frequency, spending patterns, and location mismatch
- Evaluate the model using AUC-ROC, F1-score, and confusion matrix
- Provide a well-structured GitHub repository with clear documentation
Radhika Goel
📧 Email: rgoel02005@gmail.com 🔗 LinkedIn: www.linkedin.com/in/rgoel02005