Medical text-oriented entity relationship joint extraction method

Abstract

The invention discloses a medical text-oriented entity relationship joint extraction method. The method comprises the following steps of: recombining a medical text entity relationship data set according to a triple mode; vectorization representation is carried out on the medical text statement; a parameter sharing layer is constructed by adopting a bidirectional long-short-term memory network anda self-attention mechanism; predicting a head entity label by adopting a softmax function, performing relation-tail entity joint decoding by adopting CNN-softmax, enhancing joint decoding representation by combining a head entity prediction result, and training a parameter sharing layer and a joint decoding layer by adopting a joint loss function optimization mode. According to the method, the problems that entities and relationship categories and positions in the medical text are distributed unevenly, and multiple pairs of relationships appear in the same sentence at the same time are solved, and the quality and efficiency of entity relationship extraction of the medical text can be improved.

Description

technical field [0001] The invention belongs to the technical field of medical text entity recognition, and in particular relates to a medical text-oriented entity-relationship joint extraction method. Background technique [0002] With the rapid development of natural language processing technology, especially the continuous application in vertical fields, it has greatly promoted the transition of my country's medical services from "informatization" to "intelligence". Extracting structured knowledge from free medical texts such as electronic medical records and biomedical literature is the basis for smart medical applications such as intelligent medical guidance, interrogation, and clinical decision-making assistance. It is also an important research content for building medical knowledge graphs. The joint extraction of entities and relationships is one of the core tasks of information extraction, which specifically refers to automatically identifying the location range and...

AI Technical Summary

Problems solved by technology

[0007] (1) The error transmission problem caused by the pipeline entity and relation extraction method;

[0008] (2) The information sparse problem of multiple entities and multiple relationships in the same sentence;

[0009] (3) Insufficient interaction between the entity extraction module and the relationship extraction module in the joint extraction framework

Images