Biomedical relationship extraction method based on pre-training model and self-attention mechanism

A biomedical and relationship extraction technology, applied in neural learning methods, biological neural network models, computer components, etc., can solve the problems of manual template construction and low recall rate, and achieve good results and enhance the effect of relationship extraction.

Inactive Publication Date: 2020-06-02
DALIAN UNIV OF TECH
View PDF13 Cites 10 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Traditional template and rule-based extraction methods ha

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Biomedical relationship extraction method based on pre-training model and self-attention mechanism
  • Biomedical relationship extraction method based on pre-training model and self-attention mechanism
  • Biomedical relationship extraction method based on pre-training model and self-attention mechanism

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0025] The specific implementation manners of the present invention will be further described below in conjunction with the accompanying drawings and technical solutions.

[0026] The biomedical relationship extraction method based on the pre-training model and the self-attention mechanism of the present invention first preprocesses the marked corpus to construct two biomedical entities: such as positional features between proteins, diseases, side effects, etc., and then The original corpus is converted into an input that the deep learning network model can accept; then use the ELMO pre-training model and word embedding to generate word vectors, and then connect them to generate long vectors to represent numerous sentences in biomedical texts; then pass the above long vectors through After a dropout layer, it is input to the BILSTM neural network to learn the context information of biomedical texts; through a layer of Self-attention layer, a multi-head self-attention mechanism is

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

PUM

No PUM Login to view more

Abstract

The invention belongs to the technical field of natural language processing, and relates to a biomedical relationship extraction method based on a pre-training model and a self-attention mechanism. According to the method, more complex information in the biomedical sentences can be extracted by using the ELMO pre-training model, so that the expression effect of sentence features of the biomedicaltext is improved, and the relationship between biomedical entities is better extracted; position features are added to learn the position relationship between the internal structure of the sentence and the biomedical entity, and the internal correlation between the data and the features in the sentence is better captured by using a self-attention mechanism, so that the task of biomedical relationship extraction is better completed. According to the method, the problem that only simple semantics of sentence sequences are focused in current biomedical relation extraction is solved, the relationextraction effect of regular biomedical texts can be enhanced, and the method has a good effect on relation extraction of non-regular biomedical texts on social media.

Description

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to view more

Application Information

Patent Timeline
no application Login to view more
Owner DALIAN UNIV OF TECH
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap