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

A Graph Embedding-Based Method for Quantifying Structural Differences in Networks

A network structural and quantitative method technology, applied in the field of network structural difference quantification based on graph embedding, can solve problems such as incompleteness, underestimation, and limited information, and achieve accurate quantification

Active Publication Date: 2022-06-21
HANGZHOU NORMAL UNIVERSITY
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, the above comparison methods of network structural differences can only capture part of the information of the network, and only consider the attributes of part of the description graph. The information they provide is usually limited or incomplete, and many important structural differences of the network are ignored. or underestimated

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 Graph Embedding-Based Method for Quantifying Structural Differences in Networks
  • A Graph Embedding-Based Method for Quantifying Structural Differences in Networks
  • A Graph Embedding-Based Method for Quantifying Structural Differences in Networks

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0026] The technical solutions of the present invention will be further described below with reference to the accompanying drawings.

[0027] A network structural difference quantification method based on graph embedding, the specific steps are:

[0028] Step (1). Construct the representation vector of the network node;

[0029] The graph embedding method node2vec based on random walk is used to construct the representation vector of network nodes. The complete algorithm of node2vec is as follows:

[0030]

[0031] First, a corresponding network G is constructed according to the input nodes and the edges between the nodes, G=(V, E, W). Among them, V represents a set of network nodes, E represents a set of network edges, and W represents a weight set of network edges. It is assumed that network G has N nodes, and N=100. Since node2vec is a graph embedding method with biased random walk that combines breadth-first search and depth-first search, the corresponding transition ...

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 network structural difference quantification method based on graph embedding. The method of the invention first constructs the representation vector of the network nodes and the distance vector of the network nodes, then constructs the distance distribution of the network nodes, and finally calculates the structural difference distance between the networks. The invention uses a graph embedding method such as node2vec to obtain the representation vector of the network nodes, then calculates the distance distribution of the nodes, and finally obtains the structural difference values ​​between different networks. The invention captures the global structural characteristics of the network, makes up for the defect that only part of the information in the network can be captured before, and has important significance for quantifying and identifying the structural differences between the networks.

Description

technical field [0001] The invention belongs to the field of computer technology, and relates to a network structural difference quantification method based on graph embedding. Background technique [0002] Quantifying structural differences in networks and determining isomorphisms between networks are fundamental open problems in computer science with a long history. The problem of graph isomorphism lies in judging whether two graphs are the same, thereby presenting a one-to-one correspondence between their components. Its complexity has not been defined since the mid-70s and there are currently no polynomial time algorithms, so this problem has a special place in the field of complexity theory. A recent work proposes a quasi-polynomial-time algorithm that checks whether subsections of a graph are isomorphic by a series of simple methods. However, for highly symmetric structures, computation is still very expensive. Some comparison methods based on network probability di...

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): H04L41/12H04L41/14G06Q50/00
CPCH04L41/12H04L41/14G06Q50/01
Inventor 张子柯王志鹏詹秀秀刘闯
Owner HANGZHOU NORMAL UNIVERSITY