Research On Chinese Named Entity Recognition Based On Deep Learning

Chinese named entity recognition plays a very important role in Chinese natural language processing technology.With the development of deep learning technology,Chinese named entity recognition technology is constantly improving,but the problems of word segmentation and ambiguity have not been solved.How to use deep neural networks to solve the problems in Chinese named entity recognition is a research hot topic.Aimed at the problem of Chinese named entity recognition,this thesis studies it from two aspects.The main content of the thesis includes:(1)In order to solve the problem that the character and vocabulary information is difficult to unify in the task of Chinese named entity recognition,and the existing methods are relatively complicated,this thesis proposes a Chinese named entity recognition model that integrates word segmentation features.This model strengthens the ability of vocabulary boundary understanding by introducing the auxiliary task of word segmentation and the fusion module of word segmentation features.The model achieves significant F1 value improvement on two datasets.(2)There is a certain correlation between the labels of the named entity recognition task.This thesis proposes a named entity recognition model based on label information enhancement.The model strengthens the learning of label-related features through the interaction of label semantics and character semantics.This model enhances the ability of label correlation learning by introducing label semantic interaction module and label similarity loss.The model achieves F1 value improvement on two datasets.Chinese named entity recognition is a basic but challenging task.In this thesis,two improved methods are proposed from two aspects,and a large number of experiments are carried out on multiple datasets.The experimental results show the effectiveness of the proposed method.At present,there are still many problems in Chinese named entity recognition,and researchers need to continue to work hard.