Automotive-cyber-threat-intelligence-corpus

Cyber attacks pose significant threats to connected autonomous vehicles in intelligent transportation systems. Cyber threat intelligence (CTI), which involves collecting and analyzing cyber threat information, offers a promising approach to addressing emerging vehicle cyber threats and enabling proactive security defenses. Obtaining valuable information from enormous cybersecurity data using knowledge extraction technologies to achieve CTI modeling is an effective means to ensure automotive cybersecurity. However, the lack of a specialized cybersecurity dataset for automotive CTI mining has hindered progress in this field. To address this gap, we present a novel corpus specifically designed for vehicle cybersecurity knowledge mining. This dataset, annotated using a joint labeling strategy, comprises 908 real automotive cybersecurity reports, 8195 security entities and 4852 semantic relations. Additionally, we conduct a comprehensive analysis of CTI mining algorithms based on this corpus. Our work provides a valuable resource for enhancing CTI modeling and advancing automotive cybersecurity research.

智能交通系统中的联网自动驾驶汽车面临网络攻击的严重威胁。网络威胁情报（Cyber Threat Intelligence，简称CTI），即收集与分析网络威胁信息的技术体系，可为应对新兴的车辆网络威胁、实现主动安全防御提供极具前景的解决方案。依托知识抽取技术从海量网络安全数据中获取有价值信息以完成CTI建模，是保障汽车网络安全的有效路径。然而，当前缺乏面向汽车CTI挖掘的专用网络安全数据集，这一短板已制约了该领域的研究进展。为填补这一空白，我们构建了专为汽车网络安全知识挖掘打造的新型语料库。该数据集采用联合标注策略完成标注，包含908份真实汽车网络安全报告、8195个安全实体以及4852条语义关系。此外，我们基于该语料库对CTI挖掘算法开展了全面的分析研究。本研究可为优化CTI建模以及推动汽车网络安全领域的研究发展提供宝贵资源。