要素化司法案例数据集

要素化司法案例数据集是由成都新致云服信息技术有限公司依托全国法院依法公开的裁判文书资源构建的高质量司法领域结构化数据集。数据集总规模约2TB，共包含3000+万份结构化裁判文书，覆盖19982024年民事、刑事、行政等十几类案件，数据以结构化格式存储。在数据结构上，输入端为非结构化的裁判文书原始文本，输出端为标准化司法要素标注，涵盖案情摘要、事实认定等关键字段，支持类案检索、文书生成、法学研究等智能应用。在合规性方面，本数据集经过严格清洗与脱敏处理，确保个人信息与敏感信息在生命周期内得到有效保护。

The Factorized Judicial Case Dataset is a high-quality structured judicial dataset developed by Chengdu Xinzhiyunfu Information Technology Co., Ltd. based on legally publicly available judicial document resources from national courts. The total dataset size is approximately 2 TB, containing over 30 million structured judicial documents spanning from 1998 to 2024, covering more than ten categories of cases including civil, criminal, administrative cases, and so on. The data is stored in a structured format. In terms of data structure, the input end accepts unstructured original judicial document text, while the output end provides standardized judicial factor annotations covering key fields such as case summaries, fact findings, and the like. This dataset supports intelligent applications including similar case retrieval, document generation, legal research, and other related use cases. For compliance purposes, this dataset has undergone strict cleaning and de-identification processing to ensure effective protection of personal and sensitive information throughout its entire lifecycle.

搜集汇总

背景与挑战

背景概述

该数据集是由成都新致云服信息技术有限公司构建的司法领域结构化数据资源，基于全国法院公开裁判文书，规模约2TB，包含超过3000万份覆盖民事、刑事、行政等多类案件的结构化文书。其核心是将非结构化裁判文书文本转化为标准化的司法要素标注，旨在支持类案检索、文书生成等法律智能应用，且数据经过严格脱敏处理以确保合规性。