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

Community discovery method based on nonlinear non-negative matrix factorization

A technology of non-negative matrix decomposition and community discovery, applied in the field of community discovery based on nonlinear non-negative matrix decomposition, it can solve problems such as lack of learning ability and inability to improve community discovery performance, achieving the effect of strong expressive ability and improved performance

Pending Publication Date: 2022-01-07
ZHONGKAI UNIV OF AGRI & ENG
View PDF1 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Chinese invention patent CN111400652A discloses "A Non-Negative Matrix Community Discovery Method and Movie Community Discovery Method", the disclosure date is July 10, 2020: including the following steps S11 data collection and calculation, forming similarity matrix X and L; S12 Decompose X into a non-negative matrix, X≈UV; S13 constructs an objective function O containing L; and S14 obtains an iterative formula for non-negative matrix decomposition based on the objective function, performs iterations, and completes community division. Negative matrix factorization is applied to community discovery, which improves the accuracy of the community decomposition module, but it does not have the learning ability of linear features and cannot improve the performance of community discovery

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 discovery method based on nonlinear non-negative matrix factorization
  • Community discovery method based on nonlinear non-negative matrix factorization
  • Community discovery method based on nonlinear non-negative matrix factorization

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0043] Such as figure 1 with figure 2 As shown, a community discovery method based on nonlinear non-negative matrix factorization, including the following steps:

[0044] S1: Formalize the social network and generate an undirected graph;

[0045] S2: According to the undirected graph, introduce the graph convolutional network GCN and NMF model to construct a unified loss function;

[0046] S3: Optimizing the unified loss function to obtain a local optimal solution;

[0047] S4: Extract the community discovery results according to the local optimal solution.

[0048] In the above scheme, the graph convolution network GCN and NMF model are introduced to construct a unified loss function, and joint training is performed to endow NMF with the learning ability of nonlinear features, so as to improve the performance of the NMF-based community discovery method and have strong expressive ability.

[0049] Preferably, in step S1, the undirected graph is G=(V,E), where V={v 1 ,......

Embodiment 2

[0071] figure 2 It is the overall model framework of the method, which includes an input layer, a graph convolution layer, and an output layer. The input layer uses the adjacency matrix and identity matrix of the graph as input, and the graph convolution layer uses a 2-layer graph convolution network. The activation function is ReLU , the output layer is responsible for outputting the graph node representation Z normalized by the Softmax function, and used to reconstruct the adjacency matrix of the graph. The whole model framework is trained by constructing graph reconstruction and NMF joint loss function.

specific Embodiment approach

[0074] Step 1: Formally represent the social network. Formalize the social network example as G=(V,E), where V=(v 0 ,v 1 ,v 2 ,v 3 ,v 4 ,v 5 ), E={e 01 ,e 02 ,e 12 ,e 23 ,e 34 ,e 35 ,e 45}, n=6. The corresponding adjacency matrix A and identity matrix I are:

[0075]

[0076]

[0077] Step 2: Construct the unified loss function of GCN and NMF: Here α is set to 1, and the number of communities k is set to 2.

[0078] Step 3: Optimize the solution of the unified loss function. Solve by iteratively updating the rules as follows:

[0079]

[0080] Step 4: Get community discovery results. Set the number of iterations t=10, and the result of H obtained after iteration convergence is:

[0081]

[0082] According to H, the two communities to be discovered can be directly judged c 0 and c 1 members, for example, for user node v 0 , its corresponding community affiliation strength distribution vector in H is [0.63,0.01], 0.63>0.01, so c0=c 0 ∪{v 0}, s...

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 relates to a community discovery method based on nonlinear non-negative matrix factorization, which comprises the following steps of: S1, formalizing a social network to generate an undirected graph; s2, according to the undirected graph, introducing a graph convolutional network GCN and an NMF model to construct a unified loss function; s3, optimizing the unified loss function to obtain a local optimal solution; and S4, extracting a community discovery result according to the local optimal solution. The graph convolutional network GCN and the NMF model are introduced to construct the unified loss function, and joint training is performed to endow the NMF with a nonlinear feature learning ability, so that the performance of the NMF-based community discovery method is improved.

Description

technical field [0001] The present invention relates to the field of convolutional networks, and more specifically, to a community discovery method based on nonlinear non-negative matrix factorization. Background technique [0002] Social networks (such as WeChat, QQ, and Weibo) commonly used by Internet users have a community structure. Users in the same community are more closely connected to each other, while users in different communities are less connected to each other. Community discovery on social networks can not only understand network structure characteristics, but also mine similar user groups, which has important application value for accurate recommendation of user products and services. [0003] At present, different types of community discovery methods have been proposed, among which community discovery methods based on Nonnegative matrix factorization (NMF) are more and more widely used because of their easy-to-interpret results, strong versatility, and mult...

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): G06N3/04G06N3/08G06Q50/00
CPCG06N3/08G06Q50/01G06N3/047G06N3/045
Inventor 贺超波呼增付志文郑裕龙
Owner ZHONGKAI UNIV OF AGRI & ENG
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