基于Bert的合同文本实体识别

Contract Text Entity Recognition Based on Bert

智慧合同发展的基础性任务就是对合同实体要素进行命名实体识别研究。论文通过人工订制标注规则,构建了包括合同名称、包装条款、验收条款、违约条款等23类实体的合同数据集;对于构建好的合同数据集,采用Bert-BiLSTM-CRF神经网络模型进行训练,通过对比23个合同要素的识别结果,以及与经典的神经网络模型进行对比实验结果表明,论文所用模型能够更加有效地应用于合同文本实体识别的任务。

The basic task of smart contract is to study named entity recognition of contract entity elements.In this paper,the contract data set of 23 types of entities including contract name,packaging clause,acceptance clause and breach clause is constructed by manually customizing labeling rules.For the constructed contract data set,the Bert-BiLSTM-CRF neural network model is used for training.By comparing the recognition results of 23 contract elements and comparing with the classical neural network model,the experimental results show that the model used in this paper can be more effectively applied to the task of contract text entity recognition.