AEGIS is an advanced Intrusion Detection System (IDS) framework that utilizes ensemble learning and adaptive validation techniques to provide robust network security. The framework combines enhanced feature selection, real-time adaptability, and comprehensive validation methods to maintain high accuracy in detecting network intrusions.
- Random Forest and Gradient Boosting classifiers
- Adaptive thresholding mechanisms
- Real-time concept drift monitoring
- Temporal-aware feature selection
- 99.97% detection accuracy
- Minimized false positives
- Robust performance across diverse attack scenarios
- Efficient computational resource utilization
- Time-based cross-validation
- Threshold sensitivity analysis
- Adversarial testing
- Ensemble diversity management
- Feature selection using multiple strategies:
- Random Forest Importance
- Mutual Information
- Recursive Feature Elimination
- Advanced data preprocessing pipeline
- Robust model architecture with Random Forest ensemble
- Multi-modal validation framework
- Adaptive threshold optimization
- Advanced ensemble management
- Real-time concept drift detection
# Clone the repository
git clone https://github.com/sarah-2003/AEGIS.git
# Install required packages
pip install -r requirements.txt- Python 3.8+
- NumPy
- Pandas
- Scikit-learn
- Joblib
- Matplotlib
- Seaborn
- imbalanced-learn
from aegis import EnhancedIDS, ImprovedIDS
# Initialize the IDS system
enhanced_ids = EnhancedIDS()
improved_ids = ImprovedIDS()
# Train the model
enhanced_ids.train(X_train, y_train)
# Evaluate performance
enhanced_ids.evaluate(X_test, y_test)
# Perform advanced validation
validation_results = improved_ids.time_based_validation(X, y)
threshold_results = improved_ids.threshold_sensitivity_analysis(X, y)The system employs a multi-layered approach:
- Handling missing values
- Feature scaling
- Outlier detection
- Data normalization
- Temporal-aware selection
- Multi-strategy approach
- Feature importance ranking
- Ensemble learning
- Adaptive thresholding
- Real-time monitoring
- Time-based validation
- Concept drift detection
- Adversarial testing
| Metric | Score |
|---|---|
| Accuracy | 99.97% |
| Precision | 1.00 |
| Recall | 1.00 |
| F1-Score | 1.00 |
The framework includes various visualization tools:
- Network metrics distribution
- Protocol analysis patterns
- Flow characteristics
- Anomaly patterns
- Feature importance analysis
- Fork the repository
- Create your feature branch (
git checkout -b feature/AmazingFeature) - Commit your changes (
git commit -m 'Add some AmazingFeature') - Push to the branch (
git push origin feature/AmazingFeature) - Open a Pull Request
This project is licensed under the MIT License - see the LICENSE.md file for details.
- CICIDS2017 dataset providers
- Contributors to the scikit-learn library
- Research community in network security
- Niharika Prasanna Kumar - niharikaresearch7@gmail.com
- Sarah Rukhmuddin Rajgoli - sarahrajgoli@gmail.com
Project Link: https://github.com/Sarah-2003/AEGIS