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

Dynamic community detection model based on representation learning

A technology for detecting models and communities, applied in the field of complex networks, can solve problems such as lack of modeling ability

Inactive Publication Date: 2021-02-12
TIANJIN UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

This deterministic representation lacks the ability to model node embedding uncertainty, ignoring problems with multiple sources of information (i.e., node attributes and network structure).

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
  • Dynamic community detection model based on representation learning
  • Dynamic community detection model based on representation learning
  • Dynamic community detection model based on representation learning

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention.

[0038] refer to figure 1 , a dynamic community detection model based on representation learning, including the following steps:

[0039] S1: Dynamic network definition:

[0040] (1): The dynamic network consists of continuous network snapshots G={G 1 ,G 2 ,...,G T}, where G represents the entire dynamic network, t ∈ {1, T}, network snapshot G t ={V t ,E t}, where V t and E t represent the collection of nodes and edges in the t-snapshot respectively, {V t+1 ,E t +1} and {V t ,E t}, i.e. new nodes can join in the network and create edges for existing nodes, or previous nodes can disappear from the network, on the other hand, new edges can be formed between...

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 dynamic community detection model based on representation learning. The dynamic community detection model comprises the following steps: S1, defining a dynamic network; s2, generating a VGRNN model; and S3, combining the VGRNN with a Gaussian mixture model GMM to form a VGRNNCD. According to the invention, a two-step strategy is adopted, namely, low-dimensional embedded representation of a dynamic network is obtained firstly, then clustering analysis is performed by using a clustering algorithm, migration to the dynamic network is performed, a VGRNN algorithm is used,time information is introduced into a GRNN, and the proposed VGRNNCD uses graph self-coding to replace variational graph self-coding in the VGRNN as an improvement; and then clustering analysis is performed by using a Gaussian mixture model (GMM). According to the model, the potential change of the dynamic graph can be better captured by using the high-order hidden random variable, the expressionability of the model is stronger, the fitting ability to data is stronger, and a better effect is achieved when the model is applied to execute a community detection task.

Description

technical field [0001] The invention relates to the technical field of complex networks, in particular to a dynamic community detection model based on representation learning. Background technique [0002] In recent years, community detection has become one of the most challenging problems in complex network analysis and has a wide range of applications. Community structure is an important feature of complex networks. Community detection aims to detect the real community structure in complex networks. It is conducive to mastering the internal laws and characteristics of related complex networks. Currently, most of the existing research on community detection in complex networks focuses on static networks, where many real-world problems are modeled with dynamic networks, where the network is constantly evolving over time. Such networks are typical in social networks, citation networks, and financial transaction networks. [0003] Nowadays, representation learning has been 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
IPC IPC(8): G06F16/9536G06F16/901G06F16/906G06Q50/00
CPCG06Q50/01G06F16/9024G06F16/906G06F16/9536
Inventor 张孟玄焦鹏飞王文俊潘林孙越恒
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