priamai/AnnoCTR

AnnoCTR数据集包含400份来自商业CTI供应商的网络威胁报告，这些报告描述了威胁相关的信息，如战术、技术、行为者、工具和目标行业。报告由领域专家标注了命名实体、时间表达式和网络安全特定概念，包括组织、地点、行业部门、时间表达式、代码片段、黑客组织、恶意软件、工具、战术和技术。数据集分为训练集、开发集和测试集，分别占60%、15%和25%。训练集用于模型训练，开发集用于模型选择，测试集用于评估。

The AnnoCTR dataset consists of 400 cyber threat reports from commercial CTI vendors, detailing threat-related information such as tactics, techniques, actors, tools, and targeted industries. The reports have been annotated by domain experts with named entities, temporal expressions, and cybersecurity-specific concepts, including organizations, locations, industry sectors, time expressions, code snippets, hacker groups, malware, tools, tactics, and techniques. The dataset is divided into train, dev, and test sets, accounting for 60%, 15%, and 25% of the documents, respectively, and is used for model training, selection, and evaluation.