https://github.com/dvarshith/transaction-fraud-detection
Machine Learning pipeline for financial transaction fraud detection. Incorporates SMOTE, ensemble models, neural networks.
https://github.com/dvarshith/transaction-fraud-detection
arizona-state-university autoencoder catboost data-mining ensemble-learning financial-security fraud-detection imbalanced-learning kaggle lightbgm machine-learning neural-networks python smote xgboost
Last synced: 7 months ago
JSON representation
Machine Learning pipeline for financial transaction fraud detection. Incorporates SMOTE, ensemble models, neural networks.
- Host: GitHub
- URL: https://github.com/dvarshith/transaction-fraud-detection
- Owner: dvarshith
- License: mit
- Created: 2025-02-12T05:01:55.000Z (8 months ago)
- Default Branch: main
- Last Pushed: 2025-02-12T05:21:54.000Z (8 months ago)
- Last Synced: 2025-02-12T06:30:20.228Z (8 months ago)
- Topics: arizona-state-university, autoencoder, catboost, data-mining, ensemble-learning, financial-security, fraud-detection, imbalanced-learning, kaggle, lightbgm, machine-learning, neural-networks, python, smote, xgboost
- Language: Jupyter Notebook
- Homepage:
- Size: 31.3 KB
- Stars: 0
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# Transaction Fraud Detection
[](https://www.python.org/)
[](https://scikit-learn.org/)
[](LICENSE)
## Overview
This repository contains a **fraud detection** pipeline for financial transactions, leveraging **data preprocessing**, **feature engineering**, **class imbalance handling (SMOTE)**, and a **diverse set of machine learning models** (Logistic Regression, Random Forest, LightGBM, CatBoost, XGBoost, and ensemble methods).
**Highlights**:
- **Novel feature engineering** (time-based features, transaction amount bucketing, etc.)
- **Handling imbalanced data** via SMOTE
- **Boosting algorithms** (LightGBM, XGBoost, CatBoost) for high-dimensional data
- **Advanced neural network approach** with a supervised AutoEncoder for anomaly detection
- **Stacking and voting ensembles** for robust, high AUC-ROC performance
Our best model (LightGBM) achieved **AUC-ROC of 0.89** on the Vesta Corporation dataset.
## Data
We use the **Vesta Corporation dataset** (part of a Kaggle competition) (https://www.kaggle.com/competitions/ieee-fraud-detection/overview) which includes:
- **Transaction data** (TransactionID, card info, transaction amount, time, etc.)
- **Identity data** (Device info, etc.)
**Due to size and privacy concerns,** the real dataset is **not** included in this repo.
**Key columns**:
- `TransactionID`
- `isFraud` (target)
- `TransactionDT`, `TransactionAmt`
- `Category features` (ProductCD, card1, card2, etc.)
- `Identity features` (DeviceType, DeviceInfo)
## Methodology
1. **Data Preprocessing**
- Missing value imputation
- High-correlation feature removal (via correlation heatmap)
- Encoding categorical features (one-hot or label encoding)
2. **Feature Engineering**
- **Transaction amount bucketing** (micro, small, etc.)
- **Time-based features** (day-of-week, hour-of-day)
- **Email domain grouping** (e.g., major providers vs. niche)
3. **Handling Class Imbalance**
- **SMOTE** (Synthetic Minority Oversampling Technique) to oversample the minority (fraud) class.
4. **Model Training**
- **Logistic Regression**, **Random Forest** (baselines)
- **LightGBM**, **CatBoost**, **XGBoost** (boosting methods)
- Hyperparameter tuning via Bayesian Optimization
- AUC-ROC as primary metric
5. **Ensemble Methods**
- **Voting** (soft voting across LGBM, CatBoost, XGB, etc.)
- **Stacking** with a meta-learner
6. **AutoEncoder** (Optional Neural Approach)
- A supervised autoencoder that outputs fraud probability (or uses reconstruction error).
## Results
```
|------------------|----------|
| Model | AUC-ROC |
|------------------|----------|
| Logistic Reg | 0.80 |
| Random Forest | 0.855 |
| LightGBM | **0.89** |
| CatBoost | 0.881 |
| XGBoost | 0.874 |
| Voting Ensembles | 0.86 |
| Stacking | 0.88 |
| AutoEncoder | 0.86 |
|------------------|----------|
```
**LightGBM** emerges as the top performer with **0.89** AUC-ROC, balancing speed and accuracy on this high-dimensional dataset.
## Usage
1. **Clone the repo**:
```
git clone https://github.com/YourUser/transaction-fraud-detection.git
cd transaction-fraud-detection
```
2. **Set up environment**:
```
conda create -n fraud python=3.8
conda activate fraud
pip install -r requirements.txt
```
(Create a requirements.txt if you like.)
3. **Jupyter Notebook**:
```
jupyter notebook notebooks/main.ipynb
```
Adjust paths as needed to point to your dataset.
## Next Steps
- Explore other techniques for class imbalance (e.g., ADASYN, cost-sensitive learning).
- Investigate deeper neural network architectures or specialized anomaly detection methods.
- Implement real-time streaming pipelines (Spark Streaming, Kafka) for transaction-level fraud detection.
## Acknowledgments
- Dataset by Vesta Corporation [https://www.kaggle.com/competitions/ieee-fraud-detection/overview].
- Project under Dr. Yanjie Fu, Arizona State University.
## License
This project is released under the `MIT License`. That means you’re free to use, modify, and distribute the code, but you do so at your own risk.
## Contact
Author: Varshith Dupati
GitHub: @dvarshith
Email: dvarshith942@gmail.com
Issues: Please open an issue on this repo if you have questions or find bugs.