Tamim028/Graph-Based-Intrusion-Detection-for-CAN-Bus-Security-An-Explainable-AI-Approach
GitHub: Tamim028/Graph-Based-Intrusion-Detection-for-CAN-Bus-Security-An-Explainable-AI-Approach
该框架通过图结构特征提取与可解释 AI 方法，对 CAN 总线通信数据进行入侵检测与模型解释。
Stars: 0 | Forks: 0
# 面向 CAN 总线安全的基于图的入侵检测：一种可解释 AI 方法
该研究框架包括：
1. 从数据集构建图
2. 使用“Page Rank”、“Closeness Centrality”、“Betweenness Centrality”、“In-Degree Centrality”和“Out-Degree Centrality”等指标进行特征提取
3. 通过综合指标进行模型开发与评估
4. 使用 SHAP (SHapley Additive exPlanations) 对所提出的模型进行可解释 AI (XAI) 分析
### 参考数据集
@inproceedings{rajapaksha2024can,
title={CAN-MIRGU: a comprehensive CAN bus attack dataset from moving vehicles for intrusion detection system evaluation.},
author={Rajapaksha, Sampath and Madzudzo, Garikayi and Kalutarage, Harsha and Petrovski, Andrei and Al-Kadri, M Omar},
year={2024},
organization={Network and Distributed System Security (NDSS)}
}
## 引用
如果您在研究中使用了此代码库，请引用我们的论文：
```
@inproceedings{chowdhury2025graph,
title={Graph-Based Intrusion Detection for CAN Bus Security: An Explainable AI Approach},
author={Chowdhury, Md Tamim Dari and Devnath, Maloy Kumar},
booktitle={International Conference on Data Science, AI and Applications},
pages={209--222},
year={2025},
organization={Springer}
}
```
标签：Apex, CAN总线, 可解释AI, 图神经网络, 机器学习, 车载网络安全, 逆向工具