Unlock instant, AI-driven research and patent intelligence for your innovation.

Complete entity relationship extraction method and device

An entity relationship, entity technology, applied in the field of information processing

Active Publication Date: 2020-04-14
BEIJING UNISOUND INFORMATION TECH
View PDF3 Cites 1 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The invention provides a method for extracting a complete entity relationship, which is used to solve the problem of completely extracting entities when entity overlap occurs in entity relationship extraction

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
  • Complete entity relationship extraction method and device
  • Complete entity relationship extraction method and device
  • Complete entity relationship extraction method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0085] The preferred embodiments of the present invention will be described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described here are only used to illustrate and explain the present invention, and are not intended to limit the present invention.

[0086] The following is an explanation of some of the terms involved in this application:

[0087] The BIO tag includes three tags B tag, I tag, O tag. B is the abbreviation of Begin, I is the abbreviation of Inside, O is the abbreviation of Other, the B label indicates that the current word is the beginning word of a preset entity; the I label indicates that the current word is a A preset entity (that is, any entity) is a word other than the start word; O label, the current word does not belong to the preset entity, that is, the current word belongs to a word in other entities. The three tags are collectively referred to as the BIO tag.

[0088] Entity-relationsh...

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 a complete entity relationship extraction method. The method comprises the following steps: firstly preprocessing a word sequence and then inputting to a neural network, using aneural network encoder part to encode word sequence information and transmitting the encoding result to the decoder part of the neural network to be decoded, obtaining a relation triple only containing entity start words, and obtaining the relation triple containing complete entities by integrating the relation triple with BIO labels of word sequences obtained by a classifier. The BIO-based labeland the copying mechanism are comprehensively utilized, when entity relations are overlapped, the complete entity relation can be extracted, and it can be ensured that the obtained entity relation isaccurate.

Description

technical field [0001] The invention relates to the field of information processing, in particular to a method and device for extracting a complete entity relationship. Background technique [0002] The existing neural network-based entity relationship extraction method when there is entity overlap mainly consists of the following steps: first, the word sequence is vectorized and preprocessed, and then the preprocessed word sequence is input into the encoder, and the encoder The input word sequence is encoded and an encoded representation of the input word sequence is generated. Then the decoder reads in the encoded representation, and then copies the entity of a single word to directly generate each relational triple. [0003] However, the existing methods use the copy mechanism to copy only one word from the input word sequence when generating entity relationships. However, in many cases, an entity can contain multiple words. For example, the entity Beijing Daoxiang Villa...

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): G06F16/35G06F40/279
CPCG06F16/355
Inventor 曾祥荣
Owner BEIJING UNISOUND INFORMATION TECH