Label propagation method based on propagation limitation

A label propagation and labeling technology, applied in the field of data processing, can solve the problems of different results, low application value, and poor algorithm stability, and achieve the effects of stable division results, reduced mutual interference, and accurate performance

Inactive Publication Date: 2016-06-01
XIDIAN UNIV
View PDF4 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The advantage of this algorithm is that the calculation process is very simple and the calculation speed is very fast, but the disadvantage is that the stability of the algorithm is poor, and the results obtained by each operation will be very different, resulting in low practical value

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
  • Label propagation method based on propagation limitation
  • Label propagation method based on propagation limitation
  • Label propagation method based on propagation limitation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment 1

[0031] The present invention is a method based on propagation restriction label propagation, see figure 1 , use 2-node substructure modeling for complex networks to limit the label propagation between modules, reduce the mutual interference of label propagation between modules, and divide a large and complex network into multiple modules, including the following steps:

[0032] Step 1, input a large complex network, the given large complex network uses similarity formula to generate 2-node substructure,

[0033] There are many different natural modules in large complex networks, see figure 2 (a), when initializing, define a large complex network as G: Let is an undirected and unweighted network, V is a node set, E is an edge set, and for any edge ij of G=(V,E), use the similarity formula:

[0034] S i m = | N ( v i ...

Embodiment 2

[0044] The label propagation method based on propagation restriction is the same as in embodiment 1, wherein in step 1, a similarity formula is used to generate a 2-node substructure, and the specific steps are:

[0045] 1.1. Initialize all nodes of the large complex network G(V,E) as unassigned nodes;

[0046] 1.2. Randomly select a node v that is not assigned a network i , if v i If there are unassigned neighbor nodes, choose from them with v i Vertex v with maximum vertex similarity j , put v i and v j as a 2-node substructure; otherwise v i Form a 2-node substructure by itself;

[0047] 1.3. Repeat step 2 until all nodes in the large complex network are allocated.

[0048] see figure 2 , figure 2 a is a schematic diagram of the original network data, nodes 1, 2, 3, and 4 belong to the same module, and nodes 5, 6, 7, and 8 belong to the same module. figure 2 b is the 2-node substructure divided by the present invention. In the figure, the triangle 1,3 nodes, th...

Embodiment 3

[0052] Based on the propagation restriction label propagation method, the same as in embodiment 1-2,

[0053] In this example, the module mining in the topological network of the national power grid in the western cities of the United States is completed by using the present invention. see image 3 There are 4,941 nodes and 13,188 edges in the topological network of the National Grid of western cities in the United States. There are no self-loops and repeated edges in the network, that is, there is no edge from node A to node A, and there is at most one edge between any two nodes.

[0054] see image 3 , utilize the implementation steps of the present invention as follows:

[0055] The first step is to preprocess the topological network data of the national power grid in the western cities of the United States, and obtain the network data G=(V,E);

[0056] In the second step, the network G=(V, E) is processed, and G=(V, E) is divided into 2-node substructures using the met...

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 label propagation method based on propagation limitation, and solves the problems that an original label propagation method is instable and the accuracy of a result is low. The label propagation method comprises the following steps of generating 2-node sub structures of a given complex network by using a similar formula; distributing the same labels to nodes of all 2-node sub structures; updating the labels of each node, wherein the updating principle is as follow: updating a self label according to the labels with the maximum occurrence frequency in adjacent nodes; repeatedly executing the updating step, and enabling the labels of each node not to change or reach maximum iterations; completing system updating, thus forming a module by the nodes of the same labels; dividing the complex network into multiple modules. When label propagation is carried out by taking the 2-node sub structures as initial conditions, different modules are not in mutual interference. According to the label propagation method disclosed by the invention, modules in a big complex network can be quickly and efficiently detected, and division is accurate; the label propagation method is used for carrying out module division on the complex network, so that the complex network is further analyzed.

Description

technical field [0001] The invention belongs to the technical field of data processing, and mainly relates to the analysis and processing of real network data by using computer data mining technology, in particular to a label propagation method based on propagation restriction, which is especially suitable for the analysis and processing of modules in complex networks. Background technique [0002] The real world is complex, and there are universal connections and mutual dependencies among various things. With the rapid development of computer technology, people use complex networks to model complex things and systems, and conduct in-depth research and exploration through network analysis and data mining methods to explore the correlation and laws of related factors between complex things. At the end of the 20th century, human beings made a breakthrough in the understanding of complex networks. In addition to the well-known small-world characteristics and scale-free characte...

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): G06F17/50
CPCG06F30/18
Inventor 姚勇刘慧慧刘志镜冯阿敏武文骁王炳华
Owner XIDIAN 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