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

A knowledge map-oriented relational classification method

A technology of relational classification and knowledge graph, applied in text database clustering/classification, unstructured text data retrieval, semantic tool creation, etc., can solve the problem of low classification accuracy and achieve the effect of improving accuracy

Inactive Publication Date: 2019-01-29
BEIJING UNIV OF POSTS & TELECOMM
View PDF3 Cites 3 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the current relationship classification methods all have the problem of low classification accuracy.

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
  • A knowledge map-oriented relational classification method
  • A knowledge map-oriented relational classification method
  • A knowledge map-oriented relational classification method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0027] In order to make the purpose, technical means and advantages of the present application clearer, the present application will be further described in detail below in conjunction with the accompanying drawings.

[0028] According to the applicant's observation and summary, there is often a dependency between multiple relationships corresponding to the same entity. For example, if an entity pair has the relationship "born in", then the probability of this entity pair having the relationship "nationality" is relatively high, while the probability of having the relationship "capital" is relatively low. Therefore, it is beneficial to use the dependencies between relations for relation classification. However, the current relationship classification methods do not directly describe the dependencies between relationships, so the accuracy of relationship classification is low. Based on the above analysis, this application proposes a Dependency-Aware Relation Classification (DA...

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 present application discloses a knowledge map oriented relationship classification method, comprising: a, receiving an entity pair and a sentence packet of the entity pair, determining an initialpacket representation of the sentence packet, and taking the initial packet representation as a current packet representation; B, taking the current packet representation of the sentence packet and the last output relationship as a vector, as the input of the GRU, calculating the current hidden state by the GRU; according to the current hidden state, using a classifier to calculate the probabilities of each non-output relationship under the conditions of the current output relationship and the current packet representation, and selecting the relationship with the maximum probability as the relationship of the current output; c, returning to step b until the output ends. By applying the present application, the accuracy of relationship classification can be improved.

Description

technical field [0001] The present application relates to technologies related to knowledge graphs, in particular to a relational classification method oriented to knowledge graphs. Background technique [0002] In recent years, people have built many large-scale knowledge graphs, such as Freebase, DBpedia, YAGO, and NELL. The knowledge graph abstracts the knowledge expressed by humans in the form of natural language in the real world into triple structured knowledge that is readable by both humans and machines, namely <head entity, relation, tail entity>. This knowledge has become the basis and key resource for many tasks in machine learning, natural language processing, and artificial intelligence applications, including search engines, question answering systems, voice assistants, intelligent customer service, text understanding, and machine translation. Although the existing knowledge graph has tens of millions or even billions of triplet knowledge, it still has a...

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): G06F16/35G06F16/36
Inventor 苏森程祥贾宁宁
Owner BEIJING UNIV OF POSTS & TELECOMM
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