Small sample knowledge graph completion method, system and device and storage medium

A knowledge map and small sample technology, applied in the field of graph data mining, can solve problems such as poor training effect, insufficient training data, and ignorance, and achieve the effect of enhancing expression ability, improving accuracy, and improving completion effect

Active Publication Date: 2022-06-03
UNIV OF SCI & TECH OF CHINA
View PDF4 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0012] To sum up, although the existing technical solutions have achieved certain results, they have the following difficulties for small samples: 1. The traditional knowledge map completion method needs a large number of triples as training data, but in small samples There is not enough training data in the scene, resulting in poor training effect
2. Other small-sample knowledge map completion methods ignore the semantic interaction between neighboring entities and entities, and between entities and relationships when obtaining the representation of entities

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
  • Small sample knowledge graph completion method, system and device and storage medium
  • Small sample knowledge graph completion method, system and device and storage medium
  • Small sample knowledge graph completion method, system and device and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

The embodiment of the present invention provides a small sample knowledge graph completion method, such as figure 1 As shown, it mainly includes the following steps:

Step S1, the collection, screening and preprocessing of knowledge graph data.

[0025] In the embodiment of the present invention, all small sample relationships are extracted from the knowledge graph to be completed, a support set composed of several support triples is extracted for each small sample relationship, and for each small sample relationship, a combination of a given Several query entity pairs of , construct several query triples, and extract all triples containing non-small-sample relationships as background knowledge graphs; among them, small-sample relationships appear less frequently than non-small-sample relationships, and each supporting triple The group includes a supporting entity pair and the relationship of the supporting entity pair, and the supporting entity pair and the query entity ...

Embodiment 2

The present invention also provides a small sample knowledge graph completion system, which is mainly implemented based on the method provided in the foregoing embodiment 1, such as Figure 5 As shown, the system mainly includes:

The knowledge graph data collection and preprocessing unit is used to extract all the small sample relationships from the knowledge graph to be completed, and for each small sample relationship, extract a support set composed of several support triples, and for each small sample relationship. The sample relationship combines a number of given query entity pairs to construct several query triples, and extracts all triples containing non-small-sample relationships as background knowledge graphs; among them, the number of occurrences of small-sample relationships is less than that of non-small-sample relationships. Each support triplet includes a support entity pair and a relationship between a support entity pair, and both the support entity pair and...

Embodiment 3

[0084] Further, the processing device further includes at least one input device and at least one output device; in the processing device, the processor, the memory, the input device, and the output device are connected through a bus.

[0085] In this embodiment of the present invention, the specific types of the memory, the input device, and the output device are not limited; for example:

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 small sample knowledge graph completion method, system and device and a storage medium, which can enhance the representation ability of a central entity by introducing interaction between entities and a relationship between the entities. Besides, different features reflected by different entity pairs in a support set can be captured through semantic interaction modeling, different aspects of the small sample relationship can be effectively represented, finally, the accuracy of small sample relationship connection prediction is further improved, the small sample knowledge graph completion effect is improved, and the prediction efficiency is improved. And effects in related applications such as search engines, question-answering systems, recommendation systems and the like are improved.

Description

technical field [0001] The invention relates to the field of graph data mining, in particular to a small sample knowledge graph completion method, system, device and storage medium. Background technique [0002] A graph is a data structure consisting of nodes and edges connecting the nodes. A knowledge graph is a special kind of graph. The edges in the graph have different types, and different types of edges have different semantics. In a knowledge graph, a node represents an entity and an edge represents a relationship. Entities can represent any kind of objectively existing objects or any conventional concepts in nature, while relationships are used to describe the interaction and dependencies between different objects. Knowledge graph is a way for humans to express and store world knowledge. Therefore, it has high research value and application value, and has also attracted extensive attention from academia and industry. [0003] At present, knowledge graphs are widely...

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): G06F16/36G06F40/30G06N3/04G06N3/08
CPCG06F16/367G06F40/30G06N3/08G06N3/042G06N3/048G06N3/045
Inventor 徐童陈恩红罗鹏飞朱熹
Owner UNIV OF SCI & TECH OF CHINA
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