Medicine relationship classification method based on multilayer convolutional neural network

A convolutional neural network and relationship classification technology, applied in the field of drug relationship classification, can solve problems such as poor classification effect and inaccurate text feature extraction, and achieve the effect of increasing the number of nodes, enhancing the classification ability, and improving the accuracy.

Active Publication Date: 2018-08-03
NORTHWEST UNIV(CN)
View PDF4 Cites 11 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The purpose of the present invention is to provide a drug relationship classification method based on a multi-layer convolutional neural network to solve the problem of inaccurate text feature extraction when classifying drug relationships in the prior art, resulting in poor classification results

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
  • Medicine relationship classification method based on multilayer convolutional neural network
  • Medicine relationship classification method based on multilayer convolutional neural network
  • Medicine relationship classification method based on multilayer convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] Drug relationship refers to the relationship between two drugs. For example, macrolide antibiotics such as roxithromycin, azithromycin, and clarithromycin are taken at the same time as the cardiotonic drug digoxin, which is prone to symptoms such as nausea and vomiting. , That is to say, there is a drug relationship of "inhibition" between macrolide antibiotics and digoxin. However, when a new drug is launched, it is time-consuming and labor-intensive to use human intervention to classify the relationship between the drug and other drugs. In the present invention, the original text containing multiple drug entities is input into the trained multi-layer convolutional neural network, the network model can classify the relationship between drug entities, and output the corresponding category labels, saving manpower At the same time, the number of convolutional layers in the existing network model is increased, and the number of pooling layers is correspondingly increased. ...

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 discloses a medicine relationship classification method based on a multilayer convolutional neural network. The method comprises the following steps that: collecting an original medicinetext of an English form to obtain an original medicine text set; preprocessing the original medicine text set; constructing the multilayer convolutional neural network for training; obtaining a medicine relationship classification multilayer convolutional neural network; and utilizing the medicine relationship classification multilayer convolutional neural network to carry out medicine relationship classification to obtain a medicine relationship classification result. By use of the method provided by the invention, the multilayer convolutional neural network is improved, a presentation layeris increased, the input medicine text is converted into a medicine text vector, in addition, a position vector based on a relative distance is added, so that the medicine text feature vector is moreaccurately extracted, and the accuracy of the medicine relationship classification method based on the multilayer convolutional neural network is improved.

Description

technical field [0001] The invention relates to a drug relationship classification method, in particular to a drug relationship classification method based on a multi-layer convolutional neural network. Background technique [0002] In recent years, with the rapid development of life sciences, a large amount of biomedical literature has been formed. According to statistics, only the biomedical literature database Medline contains more than 23 million biomedical literature, which contains a large amount of unstructured biomedical knowledge. Structural representation of this knowledge is helpful to the establishment of biomedical relational databases, thereby improving the efficiency and accuracy of biomedical literature retrieval, and helping researchers quickly locate target information and related literature. [0003] Biological entity relationship extraction refers to the extraction of relationships between entities from biomedical literature. Entities include proteins, ...

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
Patent Type & Authority Applications(China)
IPC IPC(8): G06F17/30
CPCG06F16/35G06F16/367
Inventor 冯筠杜晓东孙霞陈静马龙
Owner NORTHWEST UNIV(CN)
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