Entity classification method, system and computer readable storage medium

A classification method and entity technology, applied in the field of knowledge graph, can solve the problems of low classification performance and low amount of information, and achieve the effect of improving entity classification performance

Active Publication Date: 2021-05-28
IFLYTEK (SUZHOU) TECH CO LTD
View PDF5 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the traditional entity classification only focuses on coarse-grained categories such as people, organizations, and places, and the amount of information expressed is often too small, which makes the performance of entity classification low.

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 classification method, system and computer readable storage medium
  • Entity classification method, system and computer readable storage medium
  • Entity classification method, system and computer readable storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0027] Compared with coarse-grained entity classification, fine-grained entity classification forms a category path in the type tree of the knowledge base by assigning more specific categories to entities, thus having a larger amount of information and richer semantics. More effectively mine the semantic and relationship information of entities, so as to provide help for graph completion tasks such as entity linking and relationship extraction. Such as figure 1 As shown, the entity Yao Ming would be associated with the category path "thing-person-sports-person-basketball player".

[0028] At the same time, there are usually many low-frequency entities with little or no triple information in the knowledge base. Moreover, most of the current fine-grained classification models directly classify the leaf nodes of the type tree, and do not make full use of the hierarchical relationship between categories.

[0029] In addition, there is a certain association between the attributes...

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 method, system and computer-readable storage medium for entity classification, wherein the method includes the following steps: obtaining the type tree of the knowledge base, and generating a feature vector of the entity to be classified in the knowledge base, wherein the feature vector includes text features Vector, non-text feature vector and category feature vector, the category feature vector is generated based on the type tree; use the preset classification model to classify the entity to be classified according to the feature vector of the entity to be classified, so as to obtain the category path of the entity to be classified in the type tree. The entity classification method in the embodiment of the present invention utilizes the hierarchical relationship between entity categories when performing entity classification, thereby enabling layer-by-layer classification of entities to be classified from coarse-grained types to fine-grained types, improving entity classification. performance.

Description

technical field [0001] The present invention relates to the field of knowledge graphs, in particular to an entity classification method, system and computer-readable storage medium. Background technique [0002] In recent years, with the rapid development of big data and artificial intelligence, Knowledge Graph (KG) has attracted widespread attention from academia and industry with its powerful data description capabilities. [0003] The construction process of knowledge map mainly includes three steps: information extraction (entity extraction, relation extraction, attribute extraction), knowledge fusion and knowledge processing, and entity classification plays a very important role in the process of building knowledge map. However, traditional entity classification only focuses on coarse-grained categories such as people, organizations, and locations, and the amount of information expressed is often too small, which makes entity classification performance low. Contents o...

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 Patents(China)
IPC IPC(8): G06F16/35G06F16/36G06F40/295G06N3/08
CPCG06N3/084G06F16/35G06F16/367G06F40/295
Inventor 李直旭张兆银何莹徐小童
Owner IFLYTEK (SUZHOU) TECH CO LTD
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