CAKG: A Framework for Cybersecurity Threat Detection of Automotive via Knowledge Graph

With the increasing complexity and connectivity of vehicles, ensuring their security has become a critical concern. In this study, we propose CAKG: A Framework For Cybersecurity Threat Detection Of Automotive Via Knowledge Graph, achieved with a knowledge graph for vehicle vulnerability and threat intelligence. We integrate existing cyber security knowledge bases to analyze potential attack surfaces and scenarios specific to automobiles. By using keyword extraction and text similarity analysis, we identify threats relevant to the event produced by Intrusion Detection Systems(IDS) in automotive. Moreover, we leverage another knowledge graph to analyze attack logs gained from actual vehicles, enabling us to further correlate security events and product situational analysis. Our framework provides a holistic perspective on vehicle security, facilitating threat modeling and enhancing our understanding of potential attack scenarios.