Unlock instant, AI-driven research and patent intelligence for your innovation.

A multi-stage weighted network community structure detection method based on power-law function

A technology of weighted network and community structure, applied in the field of multi-stage weighted network community structure detection based on power-law function, can solve problems such as convergence of initial output values ​​of neurons

Active Publication Date: 2021-06-08
ZHEJIANG UNIVERSITY OF SCIENCE AND TECHNOLOGY
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, randomizing the initial output values ​​of each neuron does not converge to a satisfactory solution

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 multi-stage weighted network community structure detection method based on power-law function
  • A multi-stage weighted network community structure detection method based on power-law function
  • A multi-stage weighted network community structure detection method based on power-law function

Examples

Experimental program
Comparison scheme
Effect test

Embodiment

[0069] In this embodiment, a multi-stage Hopfield neural network energy function optimization method based on a power-law function is used to detect the community structure of the weighted network, which is divided into twelve steps. Workflow diagram such as figure 2 As shown, the twelve steps are described in detail below taking the weighted metabolic network as an example. The edge weight matrix file of the weighted metabolic network is stored in the external storage device of the computer.

[0070] Step 1: Read weighted network data. After the edge weight matrix of the weighted metabolic network in the external storage device of the computer is read into the memory, it can be expressed as W∈R N×N , N is the number of nodes in the weighted network. matrix element w ij Indicates the edge weight connecting node i and node j. N=453.

[0071] Step 2: Define the Hopfield neural network topology. The Hopfield neural network can be regarded as a two-dimensional grid of N×C,...

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 multi-stage weighted network community structure detection method based on a power-law function, which belongs to the field of neural networks. The present invention first designs a two-dimensional Hopfield neural network structure, and secondly sets the Hopfield neural network weight value based on the modularity function of the weighted network, that is, the energy function of the Hopfield neural network is a weighted network modularity function, and then designs multiple neural networks based on the power law function. The stage Hopfield neural network energy function optimization method optimizes the weighted network modularity function value. The experimental results show that this method can find a better modularity function value only by setting a small number of stages, so as to obtain the corresponding community structure division.

Description

technical field [0001] The invention relates to the field of neural networks, in particular to a multi-stage weighted network community structure detection method based on a power-law function. Background technique [0002] A network community is usually defined as a subset of a network node set. The connections between nodes in the set are relatively dense, and the connections between nodes in the set are relatively sparse. After all the nodes in the network are divided into corresponding communities, a community structure of the network is obtained. Network community structure detection has a wide range of applications in biological networks and social networks. For example, in protein interaction networks, protein factors with similar functions can be located through network community structure detection, which provides a basis for designing targeted drugs; in social networks, groups with similar interests can be screened out through network community structure detection...

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 Patents(China)
IPC IPC(8): G06N3/04
CPCG06N3/044
Inventor 丁进孙勇智谭平宁勇
Owner ZHEJIANG UNIVERSITY OF SCIENCE AND TECHNOLOGY