overthelex/cz-court-decisions
收藏资源简介:
捷克法院判决数据集是最大的捷克法院判决公开数据集,包含871,171个判决全文,涵盖所有法院层级。数据集分为两个配置:justice配置包含539,622个来自地区、区域和高等法院的判决(2020-2026年),许可证为CC BY 4.0(允许商业使用);czcdc配置包含331,549个来自最高法院、最高行政法院和宪法法院的判决(1993-2023年),许可证为CC BY-NC-SA 4.0(非商业使用)。数据集的独特性在于包含了540,000个来自捷克司法部API的低级法院判决,这些数据在其他HuggingFace数据集中不存在。数据模式包括ID、ECLI标识符、来源、法院名称、案件编号、判决类型、判决日期、法官、主题、关键词、引用法律条款和全文等字段。数据质量高,无重复记录。该数据集在相关论文中引入,用于评估LLM在法律推理方面的能力。
The Czech Court Judgments Dataset is the largest publicly available Czech court judgment dataset, containing 871,171 full judgment texts covering all court hierarchies. The dataset is divided into two configurations: 1. The `justice` configuration includes 539,622 judgments from district, regional, and high courts spanning 2020 to 2026, licensed under CC BY 4.0 (permitting commercial use); 2. The `czcdc` configuration includes 331,549 judgments from the Supreme Court, Supreme Administrative Court, and Constitutional Court spanning 1993 to 2023, licensed under CC BY-NC-SA 4.0 (for non-commercial use only). A distinct feature of this dataset is the inclusion of 540,000 lower court judgments sourced from the Czech Ministry of Justice API, which are not present in other Hugging Face datasets. The data schema includes fields such as ID, ECLI identifier, source, court name, case number, judgment type, judgment date, judges, subject matter, keywords, cited legal provisions, and full text. The dataset has high data quality with no duplicate records. This dataset was introduced in a relevant academic paper for evaluating the legal reasoning capabilities of Large Language Models (LLMs).




