LegalLPP

该数据集来源于广东政务服务网收集的真实世界法律条文数据，旨在预测事务相关的法律条文。数据集包含了处理非标准文本、相似事务以及缺失法律条文的详细预处理工作。数据被划分为训练集、验证集和测试集，并按特定比例分配。规模上，该数据集包含了140,482个原始事务、4,042部法律和269,053条法律条文。任务类型为法律条文预测（Lpp）。

This dataset is derived from real-world legal provision data collected from the Guangdong Government Service Network, with the objective of predicting transaction-related legal provisions. It encompasses detailed preprocessing procedures for handling non-standard texts, similar transactions, and missing legal provisions. The dataset is split into training, validation, and test sets following a specific proportional distribution. In terms of scale, this dataset contains 140,482 original transactions, 4,042 laws, and 269,053 legal provisions. The task type is Legal Provision Prediction (Lpp).