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

A knowledge base completion method based on variational interference and a tensor neural network

A neural network and knowledge base technology, applied in biological neural network models, knowledge expression, neural architecture, etc., can solve the problems of not considering the mutual information of hidden variables, ignoring prior knowledge, etc.

Active Publication Date: 2016-12-14
UNIV OF ELECTRONIC SCI & TECH OF CHINA
View PDF6 Cites 16 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] The purpose of the present invention is to solve the problem that the existing knowledge base completion technology ignores the prior knowledge of entities and relationships, and does not consider the pairwise interaction information between hidden variables, and proposes a variational inference based on Knowledge Base Completion Methods for Tensor Neural Networks

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 base completion method based on variational interference and a tensor neural network
  • A knowledge base completion method based on variational interference and a tensor neural network
  • A knowledge base completion method based on variational interference and a tensor neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0048] The embodiments of the present invention will be further described below in conjunction with the drawings.

[0049] The invention provides a knowledge base completion method based on variational inference and tensor neural network, such as figure 1 As shown, including the following steps:

[0050] S1. According to the triples in the knowledge base (e i ,e j ,r k ), construct the tensor Y.

[0051] Where e i Represents the subject in the triplet, referring to the i-th entity; e j Represents the object in the triple, refers to the jth entity, r k Represents the ralation in the triple, referring to the k-th relationship.

[0052] Assuming that the number of entities in the knowledge base is N and the number of relations is M, the constructed tensor Y∈R N×N×M , R N×N×M Is the three-dimensional real number space with dimension N×N×M; if the triples (e i ,e j ,r k ) Exists, then the element y corresponding to the subscript in each dimension of the tensor Y ijk Is 1, otherwise y ijk...

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 knowledge base completion method based on variational interference and a tensor neural network. Tensor decomposition thought and a Bayesian framework are introduced into the knowledge base completion method, and the prior knowledge of implicit variables is considered. The interaction between the implicit variables is discussed and nonlinear expression is performed through the neural network, and consideration of nondeterminacy is added. The accuracy of the knowledge base method is improved obviously, and compared with the existing technology, the knowledge base completion method has been greatly improved.

Description

Technical field [0001] The invention belongs to the technical field of knowledge base completion, and specifically relates to the design of a knowledge base completion method based on variational inference and tensor neural network. Background technique [0002] Knowledge base completion is an important research topic in the knowledge base, and it plays an important role in some applications such as question answering systems and information retrieval. The knowledge base uses triples (subject, relation, object) to represent data information, which is a kind of semantic network that reveals between entities. Although a large amount of information is stored, a considerable part of the information is missing and implicit in it Therefore, when the user searches, the corresponding answer may not be found in the knowledge base for the request that may be sent. In order to solve this problem, related researchers have proposed a series of learning algorithms based on the existing triples...

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): G06N5/02G06N3/04
CPCG06N5/022G06N3/047
Inventor 徐增林贺丽荣刘斌李广西盛泳潘王雅芳
Owner UNIV OF ELECTRONIC SCI & TECH OF CHINA
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