【摘要】 针对深度学习方法处理命名实体识别任务时,经典的词向量表示只能将其映射为单一向量,无法在上下文中表示出字的多义性这一问题,提出ERNIE-BiGRU-CRF模型,该模型通过知识增强语义表示ERNIE预训练模型增强字的语义感知表示。引入多元数据知识生成语义向量,然后将字向量输入到GRU层提取特征,最后通过CRF层得到标签序列。实验结果表明,该模型在人民日报语料库中F₁值达到了94.46%。更多还原

【Abstract】 For the problem of processing the named entity recognition in the deep learning method,traditional word embedding method map words or chars into a single vector,which can’t represent the ambiguity of the word in the context.The ERNIE-BiGRU-CRF model is proposed.The model expresses the semantic perceptual representation of the enhanced words of the ERNIE pretraining model through Enhanced Representation from kNowledge IntEgration.The multivariate data knowledge is introduced to generate the semantic vector,the word embedding is input to the GRU layer to extract features,and the label sequence is obtained through the CRF layer.The experimental results show that the F₁ value of the model in the People’s Daily corpus reaches 94.46%.更多还原