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

Community-oriented multi-layer network representation learning method

A multi-layer network, learning method technology, applied in the field of node representation learning model, can solve problems such as increasing complexity

Pending Publication Date: 2020-02-28
TIANJIN UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Adding to the complexity of the study of this problem

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
  • Community-oriented multi-layer network representation learning method
  • Community-oriented multi-layer network representation learning method
  • Community-oriented multi-layer network representation learning method

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0049]The community-oriented multi-layer network node representation learning method proposed in this paper is mainly applied to learn node representation in multi-layer heterogeneous information networks. Follow the steps described below.

[0050] Step 1: We obtain multi-layer heterogeneous information network data, such as user-artist network.

[0051] The second step: preprocessing the data, mainly including multiple single-layer networks S i , and the interlayer dependency network D ij , the single-layer network adjacency matrix S i and interlayer dependency matrix D ij , and B matrix and initial parameters α, β, λ as the input of the algorithm. Taking the user-artist network as an example, we need to get S 1 , that is, the artist network adjacency matrix, S 2 user network adjacency matrix, and D 12 User-artist relationship matrix, if there is a comment relationship between the user and the artist, then D 12 The value of the corresponding element in is 1, otherwise...

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 community-oriented multi-layer network representation learning method, which mainly comprises the following two processes of: firstly, performing node representation learningin a single-layer network through a non-negative matrix factorization method fusing modularity and first-order second-order node similarity; then, combining a plurality of single-layer network modelsinto a unified multi-layer network node expression learning model through an interlayer dependency relationship in the multi-layer network. The method studies the problem of multi-layer network noderepresentation maintaining a community structure, and uses a microstructure and a mesoscopic structure to guide the network embedding problem in a multi-layer network. Secondly, a matrix decompositionmethod is adopted, and the convergence of a final target function is proved; finally, related experiments have been carried out in the method, and the effectiveness of the method is fully proved.

Description

technical field [0001] The invention belongs to the field of social network analysis. According to the relationship between nodes in the network, the node representation modeling is carried out on the single-layer network, and then multiple single-layer networks in the multi-layer network system are modeled through the inter-layer dependencies in the multi-layer network Perform unified processing, and finally form a unified and joint node representation learning model. Background technique [0002] With the development of information technology, the popularity of various information systems makes network data ubiquitous in our daily life, such as social media networks, e-commerce networks and so on. These networks have a large number of users, complex structures, and extremely large amounts of data generated. The substantial increase in network scale and network data volume has led to the rise of network-based tasks. By performing different network learning tasks (such as n...

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): G06Q50/00
CPCG06Q50/01
Inventor 金弟王方正
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