The paper discusses the application of Graph Neural Networks (GNNs) in bolstering defensive cyber operations. With the growing complexity and evolution of cyber attacks, traditional defensive mechanisms are proving inadequate. As a solution, the authors propose the use of modern approaches like Machine Learning, specifically GNNs, to process and learn from heterogeneous cyber threat data. The paper delves into how GNNs can help break each stage of the renowned Lockheed Martin Cyber Kill Chain, thereby preventing attacks from a defensive standpoint. The authors also explore open research areas and potential areas for further improvement in this domain.
Publication date: 15 Jan 2024
Project Page: https://doi.org/XXXXXXX.XXXXXXX
Paper: https://arxiv.org/pdf/2401.05680
