Education examination-oriented cross-domain graph matching entity recognition method

The invention discloses an educational examination-oriented cross-domain graph matching entity recognition method, which comprises the following steps of: obtaining a source domain named entity recognition data set, and training a deep neural network model by using the source domain named entity recognition data set to obtain a source domain model; obtaining education examination text data with entity category labels; constructing a target domain model by using the trained source domain model; inputting the education examination text data into the source domain model and the target domain model, and respectively calculating to obtain a source domain graph and a target domain graph; and training a target domain model by using the education examination text data, the source domain graph and the target domain graph, wherein the trained target domain model can be used for education examination text entity recognition. According to the method, the advantages of transfer learning are utilized, knowledge learned by