ZySec-AI/data-extraction

该数据集是一个全面的训练语言模型的数据集，用于法律和调查背景下的问答生成、原子关系提取和意图检测。数据集包括三个主要部分：1）问题生成和原子关系数据，包含来自复杂法律文件的训练数据，如案例研究、话题覆盖等；2）意图提取训练数据，专注于提高法律执法和法律背景下的用户意图检测；3）数据格式部分详细描述了JSONL文件的结构。数据集适用于法律AI系统、调查工具、语言模型训练以及学术研究。

This dataset is a comprehensive training resource for language models, designed for question generation, atomic relations extraction, and intent detection in legal and investigative contexts. It includes three main components: 1) Question Generation and Atomic Relations data, with training data from complex legal documents such as case studies and topic coverage; 2) Intent Extractions Training Data, focused on improving user intent detection in legal enforcement and legal contexts; 3) Data Format details the structure of the JSONL files. The dataset is suitable for Legal AI Systems, Investigation Tools, Language Model Training, and Academic Research.