Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Knowledge graph completion method and system based on unstructured information

A knowledge graph, unstructured technology, applied in the field of knowledge graph completion methods and systems based on unstructured information, can solve problems such as increased data noise, decreased training accuracy, difficulty in knowledge graph completion, etc., to improve efficiency and Accuracy, improving accuracy, and improving computational efficiency

Pending Publication Date: 2022-01-14
DAREWAY SOFTWARE
View PDF0 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0005] The inventors found that if the collected unstructured data and some structured data are simply put together, it may not help the knowledge map completion task, but may increase the data noise and reduce the training accuracy, so It is not feasible to directly fuse some free texts in the Internet with some structured data; and in the Internet, there are not many free texts containing the required target entities, or there are not many structured data themselves , is relatively sparse, making it difficult to complete the knowledge map

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
  • Knowledge graph completion method and system based on unstructured information
  • Knowledge graph completion method and system based on unstructured information
  • Knowledge graph completion method and system based on unstructured information

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0036] Embodiment 1 of the present invention provides a knowledge graph completion method based on unstructured information, which utilizes the mutual cooperation of an adversarial neural network and a graph neural network to complete the final knowledge graph completion task.

[0037] As mentioned in the background technology, at present, there is a lot of data that is different from the data of the knowledge map, but there are many free texts on the Internet that store a lot of information, which can not only help the human brain to make accurate judgments, but if Making good use of it can help the machine to make some judgments, such as the completion of the knowledge map, but if this kind of free text is directly put into the data set in the knowledge map completion task, it may not be helpful, and it may even be confusing. Adding a certain amount of noise, this embodiment considers the characteristics of heterogeneity in various data, and designs an algorithm of adversaria...

Embodiment 2

[0114] Embodiment 2 of the present invention provides a knowledge graph completion system based on unstructured information, including:

[0115] The data acquisition module is configured to: acquire missing triplet data to be completed;

[0116] An entity node identification module configured to: identify entity nodes in missing triplet data;

[0117] The sentence collection module is configured to: obtain sentences associated with entity nodes, identify entity triples in the sentences, and simultaneously input the obtained sentences into the generator to generate free text data;

[0118] The confrontation training module is configured to: combine free text data and structured text data to train the generator, and the discriminator distinguishes the prediction result of the generator's entity triples according to the entity triples in the sentence, and performs the generator and Adversarial training of the discriminator;

[0119] The knowledge map completion module is config...

Embodiment 3

[0122] Embodiment 3 of the present invention provides a computer-readable storage medium, on which a program is stored, and when the program is executed by a processor, the knowledge graph completion method based on unstructured information as described in Embodiment 1 of the present invention is implemented. A step of.

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 knowledge graph completion method and system based on unstructured information. The method comprises the following steps: identifying entity nodes in missing triple data; obtaining sentences associated with the entity nodes, identifying entity triples in the sentences, and inputting the obtained sentences into a generator to generate free text data; combining the free text data and the structured text data to train a generator, enabling that a discriminator discriminates an entity triple prediction result of the generator according to the entity triple in the sentence, and carrying out adversarial training of the generator and the discriminator; and when the discriminator succeeds in discriminating, adding the entity triples in the sentences into the knowledge graph, scoring through a graph neural network, obtaining the ranking result of the entity triples in combination with the scoring result and the previous node information of the known entity nodes lacking the triples, and completing the knowledge graph. According to the invention, the efficiency and accuracy of knowledge graph completion are improved.

Description

technical field [0001] The present invention relates to the technical field of knowledge graph completion processing, in particular to a knowledge graph completion method and system based on unstructured information. Background technique [0002] The statements in this section merely provide background art related to the present invention and do not necessarily constitute prior art. [0003] The essence of knowledge graph is a semantic web, and knowledge graph has developed rapidly in recent years and has been widely used in real life, such as factual question answering systems based on large knowledge bases such as DBpedia, YAGO, and Freebase. However, knowledge in real life is always linked one after another, so a lot of knowledge is not available in the original knowledge base, and it can also be said to be incomplete, so the task of knowledge map completion is slowly being paid attention to by people. Generally speaking, the knowledge map is composed of three parts: the...

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): G06F16/35G06F16/28
CPCG06F16/35G06F16/288
Inventor 史玉良吕梁纪风坡管永明张晖刘智勇
Owner DAREWAY SOFTWARE
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com