NER dataset related to legal texts

The following data pertains to Named Entity Recognition for legal judgment documents related to the crime of assisting in information network crimes. The dataset consists of a total of 4,236 samples, including both training and validation data, with a total of 8 labels. The file train1.json contains the raw data in JSON format, which is not divided into training and validation sets. The ner_data folder contains processed data in .txt file format, with the dataset split into training and validation sets at a ratio of 5:1. This folder also includes all label names. Ultimately, the model is trained using the processed dataset.本数据集面向帮助信息网络犯罪活动罪相关裁判文书的命名实体识别（Named Entity Recognition，NER）任务。数据集总计包含4236条样本，涵盖训练与验证数据，共设有8个标注标签。其中train1.json为未划分训练集与验证集的原始JSON格式数据；ner_data文件夹内存储了处理后的.txt格式文件，该数据集按5:1的比例划分为训练集与验证集，且文件夹内包含全部标签名称。最终将采用此处理后的数据集开展模型训练。