Entity relationship extraction method based on attention contribution degree and application thereof

An entity relationship and entity extraction technology, applied in the field of knowledge extraction, can solve the problems of entity nesting, relationship overlap, insufficient information mining in specific fields, etc., and achieve the effect of strengthening mining.

Pending Publication Date: 2022-04-29
XIANGTAN UNIV +1
View PDF0 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Among them, the research on entity relationship extraction mainly includes sequence labeling schemes and span-based schemes, but the current resea

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
  • Entity relationship extraction method based on attention contribution degree and application thereof
  • Entity relationship extraction method based on attention contribution degree and application thereof
  • Entity relationship extraction method based on attention contribution degree and application thereof

Examples

Experimental program
Comparison scheme
Effect test

Example Embodiment

[0100] Example 1

[0101] An entity relation extraction method based on attention contribution, the method includes the following steps:

[0102] S0: selected data set D';

[0103] S1: Segment the original sentence through spacy, and store the word list and the labels contained in the dataset D' into the input dataset D in the form of a dictionary

[0104] S2: Sampling the dataset D to obtain each sentence d in the input dataset D i (d i ∈ D) entity sample set and relation sample set

[0105] S3: Build a span-based entity relationship extraction model, including BERT pre-training module, entity extraction module and relationship classification module;

[0106] S4: According to each entity sample set Calculate each sentence d i each entity in eigenvector of and entity attention contribution the entity feature vector and attention contribution Combined, passed into the entity extraction module to get the predicted entity type entity of the entity extraction ...

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 provides an attention contribution degree-based entity relationship extraction method, which comprises the following steps of: performing word segmentation on an original sentence in a data set through spaCy to obtain a word list, and storing the word list and a label into an input data set D in a dictionary form; sampling the input data set D to obtain an entity sample set and a relation sample set of each sentence in the input data set D; a BERT model pre-trained on a large-scale biomedical corpus, a judicial database and a tourism database is selected, interaction information between entities is calculated through an attention contribution degree algorithm, then the interaction information is transmitted to downstream entity extraction and relation extraction tasks, and a span-based entity relation extraction model is formed; and finally, putting the entity sample set and the relationship sample set into a span-based entity relationship extraction model for training, thereby greatly improving the F1 value of an entity extraction task and the F1 value of a relationship extraction task.

Description

technical field [0001] The invention relates to the field of knowledge extraction, in particular to an entity relationship extraction method based on attention contribution, and the method is used to analyze medical reports. Background technique [0002] In the field of natural language processing, information extraction has always attracted people's attention. Information extraction mainly includes three sub-tasks: entity extraction, relation extraction and event extraction, and relation extraction is the core task and important link in the field of information extraction. The main goal of entity relationship extraction is to identify and determine the specific relationship between entity pairs from natural language texts, which provides basic support for intelligent retrieval, semantic analysis, etc., helps to improve search efficiency, and promotes the automatic construction of knowledge bases . [0003] At first, the entity relationship extraction involved in the MUC an...

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
IPC IPC(8): G06F40/284G06F40/242G06N20/00
CPCG06F40/284G06F40/242G06N20/00
Inventor 欧阳建权张晶李波
Owner XIANGTAN UNIV
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