Deep Learning Ensemble for Credit Card Fraud Detection
Developed an ensemble model using LSTM, GRU, and MLP techniques, successfully improving fraud detection rates while minimizing false positives.
- Applied data resampling and preprocessing for imbalanced datasets
- Achieved significant improvement in detection rate with low false positive rates
- Built an interactive Streamlit web application for real-time predictions
Python
TensorFlow
LSTM
GRU
Streamlit