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

Deep community discovery method fusing node attributes

A technology for community discovery and fusion of nodes, applied in the field of graph segmentation, can solve problems such as ignoring attribute information

Inactive Publication Date: 2021-09-17
ARMY ENG UNIV OF PLA
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

Most algorithms only utilize topological information, while ignoring important attribute information

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
  • Deep community discovery method fusing node attributes
  • Deep community discovery method fusing node attributes
  • Deep community discovery method fusing node attributes

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0032] The goal of an autoencoder is to reconstruct the original input so that the output is as close as possible to the input. In this way, the output of the hidden layer can be regarded as a low-dimensional representation of the original data, so as to extract the features contained in the original data to the greatest extent. An autoencoder consists of two symmetrical components: an encoder and a decoder. A basic autoencoder can be seen as a three-layer neural network consisting of an input layer, a hidden layer, and an output layer.

[0033] Given an input data x i , the encoder converts the original data x i Mapping to the output encoding h of the hidden layer i , h i can be seen as x i The low-dimensional embedding representation of :

[0034] h i =σ(W (1) x i +b (1) ) (1)

[0035] The decoder then reconstructs the input data, is the reconstructed output data:

[0036]

[0037] After the input data is encoded and decoded, a reconstructed representation o...

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 deep community discovery method fusing node attributes, and relates to the technical field of mapping segmentation problems. The method includes: constructing a modularity matrice, wherein larger the value of the modularity is, the clearer the community structure is, the better the division of the community is, and the community structure of the network can be obtained by maximizing the modularity; constructing a deep auto-encoder capture network structure: reconstructing a modularity matrix, storing a nonlinear community structure of the network in the output H of the last layer of the hidden layer; combining node attribute information; when nodes with the same attribute are divided into different communities, executing a punishment, and carrying out community discovery by fusing link relation data and node content data at the same time. According to the method, a nonlinear structure is mined by using the deep neural network, and a more accurate community structure is obtained in combination with node attribute information.

Description

technical field [0001] The invention relates to the technical field of graph segmentation problems, in particular to a deep community discovery method that integrates node attributes. Background technique [0002] Community structure is an important structural feature that widely exists in the network. The nodes within the community are closely connected, and the nodes between communities are sparsely connected. Community discovery is the process of mining the hidden community structure in network data from a mesoscopic perspective by analyzing the interaction and potential information between nodes in the network. Community discovery provides an effective tool for exploring the potential characteristics of complex networks, and has important theoretical and practical significance for understanding network organizational structures, analyzing potential characteristics of networks, and discovering hidden laws and interaction patterns of networks. Node attributes are importan...

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): G06Q50/00G06K9/62G06N3/04G06N3/08
CPCG06Q50/01G06N3/04G06N3/08G06F18/22
Inventor 潘志松潘雨胡谷雨张磊段晔鑫张武胡亚豪丁钰
Owner ARMY ENG UNIV OF PLA