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

Interpretable link prediction method for knowledge hypergraph

A prediction method and knowledge technology, applied in the field of knowledge hypergraph, to achieve the effect of improving performance and good link prediction

Pending Publication Date: 2022-07-22
TIANJIN UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] In view of the above existing technologies, with the development of knowledge hypergraph, it has interpretability, knowledge hypergraph representation learning integrated with domain knowledge is an academic frontier issue of artificial intelligence, and has very high academic value and potential application value

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
  • Interpretable link prediction method for knowledge hypergraph
  • Interpretable link prediction method for knowledge hypergraph
  • Interpretable link prediction method for knowledge hypergraph

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0047] The interpretable knowledge hypergraph constructed in the present invention represents the learning model, such as figure 1 As shown, an example graph of a knowledge hypergraph, related logic rules and a Markov blanket is given. Among them, the circle represents an entity, the ellipse in the upper figure represents several n-tuples of the knowledge hypergraph, the ellipse in the lower figure represents a group, and the n-tuple of the knowledge hypergraph consists of multiple entities, where parentsOf represents the parent-child relationship, and couple represents The relationship between husband and wife, wasBornIn represents the relationship of birthplace, and workAsFor represents the relationship of work status, which consists of 3, 2, 3, and 3 entities respectively. The rules based on these predicates are shown in the Markov logic network in the middle map. Markov A logical network consists of logical rules and rule weights. In the variational step E, the knowledge ...

Embodiment 2

[0049] Use the interpretable connection prediction method proposed by the present invention to realize the process of knowledge hypergraph representation learning, such as figure 2 As shown, the process of the knowledge hypergraph representation learning method is as follows:

[0050] 1. Establish an interpretable knowledge hypergraph representation learning model

[0051] 1-1) Input tuples and regular data, use Markov to build joint probabilities and maximize the log-likelihood of observable tuples.

[0052] Representing the knowledge hypergraph as in ε, and A finite set of knowledge hypergraphs, entities, relations, and observable tuples, respectively. Observable tuple t i =r(e 1 , e 2 ,...,e k ), where r is a relation, each is an entity, k is the non-negative integer number of relation r, and i is the observable tuple index. each tuple t i with an indicator variable association, indicates that the tuple is true, then indicates that the tuple is false...

Embodiment 3

[0093] The method of the present invention and the related methods in the prior art are the experimental results of link prediction on the knowledge hypergraph data set or knowledge map data set.

[0094] The present invention performs link prediction on knowledge hypergraph data sets JF17K, M-FB15K, FB-AUTO and knowledge graph data sets FB15k, WN18, FB15k-237, and WN18RR.

[0095] see image 3 , on the knowledge hypergraph dataset, m-CP is selected as the embedding model of the knowledge hypergraph. Compared with pure m-CP and other knowledge hypergraph embedding methods, the technical solution of the present invention achieves better performance on almost all evaluation indicators. The reason is that the knowledge hypergraph embedding method only utilizes the semantic information in the embedding space, while the present invention uses the embedding further combined with the domain knowledge in the logic rules to predict the establishment probability value of the hidden tup...

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 an interpretable link prediction method for a knowledge hypergraph. The method comprises the following steps: constructing an interpretable knowledge hypergraph representation learning model based on a knowledge hypergraph embedding model and a Markov logic network; establishing a joint probability for all observable tuples and hidden tuples of the knowledge hypergraph through a Markov logic network, and maximizing the log likelihood of the observable tuples as a training target; optimizing a confidence lower bound of a log-likelihood function by adopting a variational EM algorithm to realize training and verification of the model; and performing link prediction on the knowledge hypergraph data set by using the verified interpretable knowledge hypergraph representation learning model, namely, taking one hidden tuple in the knowledge hypergraph data set as the input of the model, and outputting a probability value that the hidden tuple is established and the contribution degree of entities and relationships connected with the hidden tuple to the establishment of the hidden tuple by the model. By means of the method, the domain knowledge in the logic rule and semantic information in the vector space can be fully utilized, and the knowledge hypergraph representation learning effect is improved.

Description

technical field [0001] The invention relates to a knowledge hypergraph, in particular to a representation learning oriented to a large-scale knowledge hypergraph. Background technique [0002] With the rapid development of the Internet, the amount of data has exploded. In order to deeply understand the semantic information behind user queries, and then enhance the search quality of search engines, Google first proposed the concept of Knowledge Graph in 2012. A knowledge graph formally describes things in the real world and their relationships with each other, and is a large-scale semantic network that stores human knowledge in the form of graphs. It represents knowledge as a triple p(s,o), where p is the predicate, s is the subject, and o is the object. A triple p(s, o) is used to indicate that there is a relationship p between resource s and resource o, or resource s has attribute p and its value is o. [0003] Knowledge Hypergraph is a graph-structured knowledge base th...

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): G06F16/955G06F16/36G06N7/00
CPCG06F16/9558G06F16/367G06N7/01
Inventor 王鑫陈子睿王晨旭刘鑫
Owner TIANJIN UNIV
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