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

Method for establishing dynamic network model using deep convolutional neural network

A dynamic network, deep convolution technology, applied in the field of network science research, can solve the problem of low universality

Inactive Publication Date: 2017-12-01
NANCHANG HANGKONG UNIVERSITY
View PDF0 Cites 27 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, these methods require professional experience in related fields, and cannot spontaneously find a suitable representation method for network structural features in real network data in different scenarios, and their universality is not high.

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
  • Method for establishing dynamic network model using deep convolutional neural network
  • Method for establishing dynamic network model using deep convolutional neural network
  • Method for establishing dynamic network model using deep convolutional neural network

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0031] The present invention is a method for extracting high-order features of dynamic network structures using deep convolutional models. The topological structure of the dynamic network has always been affected by complex and changeable environmental factors, and the deep convolution model aims to accurately grasp the potential relationship between various variables and factors to effectively analyze the law of dynamic changes in the network structure. Further description will be given below in conjunction with the accompanying drawings and specific implementation methods.

[0032] like Figure 1 to Figure 3 As shown, G= is defined as a dynamic network, where N is a node set, E is an edge set, and the set S={G 1 , G 2 ,...,G t} is defined as a timing subgraph of a dynamic network, representing the evolution of the network over time. In the present invention, the spontaneous network analysis process is realized by establishing a deep learning model, and the specific steps...

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 method for establishing a dynamic network model using a deep convolutional neural network. On the basis of properties of the time sequence and sociality of node movement in a dynamic network, a method for extracting network structure features by using a convolution neural network based on digging of a relationship between network topology and a link state. According to the method disclosed by the invention, slicing up is carried out on dynamic time sequence data based on a chaotic time series theory; segmented network topology information is expressed by a state matrix; the state matrix is transformed into an observation map; and then high-order features that can affect the link change in the observation map are extracted by using a deep convolution neural network. Therefore, with full consideration of a complicated relationship between nodes in the dynamic network, a rule of a network structure with a time change is learned from lots of time sequence data by using a deep learning model, so that the potential mode under the dynamic network topological change is extracted effectively; and the certain support is provided for related studies like the structure evolution analysis of the dynamic network.

Description

technical field [0001] The invention belongs to the field of network science research, and mainly relates to research directions such as deep learning, chaos theory, and dynamic network analysis. Background technique [0002] According to the speed of network topology changes, traditional networks can usually be classified into three categories: static networks, steady-state networks, and dynamic networks. The state of the nodes in the static network will not change basically, and its topology has the highest stability, such as: computer network; compared with the static network, the stability of the topology of the steady state network is not high, but the change of the state of the nodes is relatively stable , such as: mobile social network, wireless sensor network, etc.; dynamic network belongs to the key research category of complex network. Compared with the former two, nodes in this type of network move frequently, and the network structure has high dynamics and comple...

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): G06N3/04G06N3/08
CPCG06N3/0418G06N3/08G06N3/045
Inventor 舒坚张学佩蔡许林刘琳岚
Owner NANCHANG HANGKONG UNIVERSITY
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