How sifat192/hybrid-network-intrusion-detection-ml-qml works
This repository is a notebook-based network intrusion detection study comparing classical machine learning with a Qiskit quantum classifier. Its documented workflow integrates traffic datasets, prepares and analyzes features, trains models, and compares evaluation results. The README supports the workflow and capabilities shown here; the notebook is the only listed implementation file, and its cells were not provided for inspection, so internal call relationships cannot be verified.
Main entry points: Hybrid_Network_Intrusion_Detection_ML_QML.ipynb Feature scaling
Components 14
Data preparation
- Data integration
Hybrid_Network_Intrusion_Detection_ML_QML.ipynb - Exploratory analysis
Hybrid_Network_Intrusion_Detection_ML_QML.ipynb - Cleaning and preprocessing
Hybrid_Network_Intrusion_Detection_ML_QML.ipynb - Feature engineering
Hybrid_Network_Intrusion_Detection_ML_QML.ipynb - Feature scaling
Hybrid_Network_Intrusion_Detection_ML_QML.ipynb
Classical learning
- Classical classifiers
Hybrid_Network_Intrusion_Detection_ML_QML.ipynb - Feature selection
Hybrid_Network_Intrusion_Detection_ML_QML.ipynb - PCA reduction
Hybrid_Network_Intrusion_Detection_ML_QML.ipynb
Quantum learning
- Quantum feature map
Hybrid_Network_Intrusion_Detection_ML_QML.ipynb - Fidelity quantum kernel
Hybrid_Network_Intrusion_Detection_ML_QML.ipynb - Quantum QSVC
Hybrid_Network_Intrusion_Detection_ML_QML.ipynb
Evaluation
- Performance comparison
Hybrid_Network_Intrusion_Detection_ML_QML.ipynb - Classification metrics
Hybrid_Network_Intrusion_Detection_ML_QML.ipynb
Other components
- Traffic CSVs
Connections 15
- Traffic CSVs → Data integration: integrate
- Data integration → Exploratory analysis: analyze
- Exploratory analysis → Cleaning and preprocessing: inform cleaning
- Cleaning and preprocessing → Feature engineering: prepare features
- Feature engineering → Feature scaling: scale features
- Feature scaling → Classical classifiers: train models
- Feature scaling → Feature selection: select features
- Feature scaling → PCA reduction: reduce dimensions
- Feature selection → Quantum feature map: encode features
- PCA reduction → Quantum feature map: encode features
- Quantum feature map → Fidelity quantum kernel: compute kernel
- Fidelity quantum kernel → Quantum QSVC: fit classifier
- Classical classifiers → Performance comparison: evaluate
- Quantum QSVC → Performance comparison: evaluate
- Performance comparison → Classification metrics: report metrics