日本侵权案件数据集（JTD）

日本侵权案件数据集（JTD）是首个针对日本法律判决预测（LJP）的数据集，专注于侵权预测及其理由提取。该数据集基于41位法律专家标注的3,477份日本民法判决，共包含7,978个实例和59,697条当事人主张。JTD数据集的创建旨在通过机器学习技术辅助法律程序，提高法律服务的效率和可及性，特别是在预测诉讼结果方面。数据集的应用领域包括法律专业人士和公众的法律咨询辅助，以及法律判决预测研究。

The Japanese Tort Dispute Dataset (JTD) is the first dataset dedicated to Japanese Legal Judgment Prediction (LJP), focusing on tort prediction and the extraction of its justifications. This dataset is constructed based on 3,477 Japanese civil law judgments annotated by 41 legal experts, containing a total of 7,978 instances and 59,697 party allegations. The JTD was developed to assist legal proceedings through machine learning technologies, enhancing the efficiency and accessibility of legal services, especially in predicting litigation outcomes. The application scenarios of this dataset include legal consultation assistance for legal professionals and the general public, as well as research on legal judgment prediction.