Electronic medical record entity relationship extraction method based on shortest dependency subtree

Abstract

The invention provides an electronic medical record entity relationship extraction method based on a shortest dependency subtree. The method comprises the following steps: firstly, extracting an entity-based shortest subtree from an original sentence through dependency syntactic analysis to compress the sentence length; secondly, coding the statements through a bidirectional long short-term memory(BLSTM) neural network, and then coding the statements through the BLSTM neural network; learning final semantic representation of the sentences through a maximum pooling layer (Max Pooling), and finally classifying the sentences through a softmax classifier to obtain an entity relationship. According to the method, noise vocabularies and compressed statement lengths can be deleted. Meanwhile, the key words representing the relations between the entities are completely reserved, so that the compressed statement semantic relations are clearer. The problem that semantic information of statements cannot be well represented due to too long statements of an existing electronic medical record entity relation extraction model is solved, and the performance of the relation extraction model is improved.

Description

technical field [0001] The invention belongs to the field of natural language processing and relates to an entity relationship extraction method, in particular to an electronic medical record entity relationship extraction method based on the shortest dependent subtree. Background technique [0002] With the advent of the big data era, data in various fields is growing rapidly. Especially for the medical field, a large number of electronic medical records are generated in clinical diagnosis and treatment, which contain a large amount of unstructured text and medical and health knowledge. Effective mining and utilization of this knowledge is of great significance to the development of medical and health care. An effective way to mine knowledge in electronic medical records is the technology related to information extraction, and the relationship extraction between conceptual entities is an important part of information extraction. [0003] Currently, there are mainly machin...

AI Technical Summary

Problems solved by technology

[0004] Aiming at the current situation and existing problems of the above-mentioned existing electronic medical record entity relationship extraction model, the present invention proposes an electronic medical record entity relationship extraction method based on the shortest dependent subtree, which overcomes the problem caused by the existing electronic medical record entity relationship extraction model due to too long sentences. The problem of not being able to represent the semantic information of the sentence well

[0006] The present invention can not only delete noise vocabulary and compress the sentence length, but also completely retain the key words representing the relationship between entities, so that the semantic relationship of the compressed sentence is more clear, so it overcomes the problem of the existing electronic medical record entity relationship extraction model. The problem of not being able to represent the semantic information of the sentence well due to the long length has improved the performance of the model

Images