LSTM-GCN-Anomaly-Detection

Hybrid approach combining LSTM and Graph Convolutional Networks (GCN) for real-time anomaly detection in dynamic computer network data.

• Kaya, Muhammed Onur, et al. “A New Hybrid Approach Combining GCN and LSTM for Real-Time Anomaly Detection from Dynamic Computer Network Data” Computer Networks, vol. 268, 2025, p. 111372, https://doi.org/10.1016/j.comnet.2025.111372.
• Author: Muhammed Onur Kaya, Mehmet Ozdem, and Resul Das

Flowchart of the proposed method:

Setup Environment

# Make sure your PIP is up to date
pip install -U pip wheel setuptools

Contributing

• If you encounter any issues with the project, feel free to report them on GitHub Issues.
• If you wish to contribute to the project, you can submit a Pull Request.

License

Our project is licensed under the GPU License.