Multilayer network clustering method based on semi-supervision

A multi-layer network and clustering method technology, applied in the field of multi-layer network clustering based on semi-supervised, can solve problems such as a lot of noise and sparse structure, and achieve the effect of effective mining and high accuracy

Inactive Publication Date: 2021-04-30
NORTHWESTERN POLYTECHNICAL UNIV
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0006] The purpose of the present invention is to provide a kind of multi-layer network clustering method based on

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
  • Multilayer network clustering method based on semi-supervision
  • Multilayer network clustering method based on semi-supervision
  • Multilayer network clustering method based on semi-supervision

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0087] The internal density function f(D) in the step S3 is defined as:

[0088]

[0089] Among them, m D Indicates the number of edges actually contained in all nodes in the dense subgraph D; n D (n D -1) / 2 represents the total number of edges owned by the dense subgraph D.

Embodiment 2

[0091] In the fusion optimization process of the consensus matrix H in step S6, the consensus prior matrix P is continuously used to minimize the dissimilarity of the row vectors corresponding to the nodes in H that have a high probability of belonging to the same cluster, and these nodes will eventually be Divided into the same cluster, for node υ i and υ j The similarity of , measured using the square of the Euclidean distance, node υ i and υ j The corresponding row vector h of row i and row j in H i and h j The calculation method of the similarity between them is:

[0092]

[0093] Among them, h i and h j represent the row vectors of the i-th row and the j-th row in the consensus matrix H respectively; υ i and υ j The more similar, then S(h i ,h j ) the smaller the value.

Embodiment 3

[0095] The specific encoding method in the step S7 is as follows:

[0096]

[0097] where P is a binary diagonal matrix, p ij =1 means υ i and υ j The probability of being divided into the same cluster increases; H is the consensus low-dimensional representation matrix of the multi-layer network; S(h i ,h j ) is used to measure h i and h j the similarity of is a diagonal matrix whose diagonal entries are L=D-P is defined as the graph regularization matrix of P; Tr(·) denotes the trace of the matrix.

[0098] refer to figure 2 , shows a detailed diagram of the method framework of the present invention. Specifically, the method (S-MCGR) proposed by the present invention mainly includes three core components. (1) Use the greedy search algorithm to obtain the consensus prior information of the multi-layer network and initialize each network layer; (2) Encode the obtained consensus prior information into the regularization term of the consensus subspace graph; (3) B...

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 multilayer network clustering method based on semi-supervision, relates to the technical field of artificial intelligence and complex networks, and not only takes obtained consensus prior information as a preprocessing means to enable a low-dimensional representation matrix H (i) of each layer obtained through symmetric non-negative matrix factorization to be more excellent. Moreover, the obtained consensus prior information is coded into a consensus subspace graph regularization item, and the consensus low-dimensional subspace H is optimized during the overall non-negative matrix factorization, so the method can make full use of the complementary topological structure information of each network layer, and can also make full use of the obtained consensus prior information; and the method is especially suitable for a multi-layer network with a large amount of noise and a sparse structure. The method is applied to social networks, protein networks and other multi-layer networks, cluster structures of various types of multi-layer networks are successfully recognized, and the method has great significance in understanding some social interaction behaviors among people, recognizing crowds with specific social attributes and improving social cooperation efficiency.

Description

technical field [0001] The invention relates to the technical fields of artificial intelligence and complex networks, in particular to a semi-supervised-based multi-layer network clustering method. Background technique [0002] By constructing a multi-layer network to describe and analyze increasingly complex social relationships, we can effectively understand the complex relationships between social users. Through the analysis and mining of the hidden public cluster structure in the multi-layer social network, it plays an important role in in-depth understanding of people's social interaction behavior on different social platforms, and it is important to comprehensively mine people with specific social attributes on different platforms and improve social cooperation efficiency. significance. In addition, using multi-layer network clustering technology can also understand the topology of the network and solve many important tasks, such as predicting the interaction relation...

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): G06K9/62
CPCG06F18/23213G06F18/2133G06F18/2155G06F18/251
Inventor 高超王震刘兴建刘晨李向华朱培灿李学龙
Owner NORTHWESTERN POLYTECHNICAL UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Try Eureka
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap