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

Unsupervised graph representation learning method and device on large-scale attribute graph based on sub-graph sampling

A large-scale attribute and learning method technology, applied in the information field, can solve the problems of limited scalability, lack of node attribute information and utilization of high-level structure information on the graph, so as to improve scalability, enhance scalability, Node vectors representing valid effects

Active Publication Date: 2020-11-17
PEKING UNIV
View PDF5 Cites 14 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

The algorithm based on the graph neural network can simultaneously utilize the structural information and node attribute information in the graph, but it is only applicable to small graphs (thousands of nodes, tens of thousands of edges), and the scalability is limited.
[0006] From the perspective of information utilization, the existing algorithms, that is, the above two types of algorithms, only focus on the local structural information in the graph in the loss function, and reconstruct edges or local neighbor nodes, lacking information on node attributes and high-level information on the graph. Utilization of hierarchical structure information (such as community)

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
  • Unsupervised graph representation learning method and device on large-scale attribute graph based on sub-graph sampling
  • Unsupervised graph representation learning method and device on large-scale attribute graph based on sub-graph sampling

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0037]In order to make the above objects, features and advantages of the present invention more comprehensible, the present invention will be further described in detail below through specific embodiments and accompanying drawings.

[0038] This patent uses a graph neural network to learn a low-dimensional vector representation of nodes in an attribute graph under an unsupervised setting. In order to improve the scalability of the algorithm, for large graphs, this patent uses the method of subgraph sampling to reduce the scale of training data. The subgraph sampling method comprehensively considers the structural information and node attribute information of the graph, making the sampled subgraph more efficient. Reasonable. In order for the network to comprehensively utilize the structural information of the graph, the node attribute information, and the community information on the graph during the learning process, a loss function related to the above three information is de...

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 an unsupervised graph representation learning method and device on a large-scale attribute graph based on subgraph sampling. The method comprises the steps of performing sub-graph sampling on an attribute graph according to structure information and node attribute information of the attribute graph to generate a plurality of sub-graphs; and performing graph learning on anauto-encoder on each sub-graph by utilizing the structure information, the node attribute information and the community information of the attribute graph to obtain low-dimensional vector representation of nodes in the attribute graph. The graph auto-encoder comprises an encoder and a decoder; the encoder adopts a graph convolutional neural network; the decoder includes a graph structure loss reconstruction decoder, a graph content loss reconstruction decoder, and a graph community loss reconstruction decoder. A user is supported to learn low-dimensional vector representations of nodes in a large-scale attribute graph in an unsupervised mode, topological structure information and node attribute information on the graph can be reserved as much as possible through the vector representations,and the vectors serve as input to be applied to different downstream tasks to conduct data mining tasks on the graph.

Description

technical field [0001] The invention belongs to the field of information technology, and in particular relates to an unsupervised graph representation learning method and device on a large-scale attribute graph based on subgraph sampling. Background technique [0002] In recent years, with the rapid development of the Internet, the scale of data has exploded, and the connection between data has become more and more complex. Graphs describe things and the relationship between things in the form of points and edges. It is a data structure that can intuitively describe the objective world. It exists widely in production and life, such as social networks, traffic road networks, and e-commerce. The nodes in these graphs usually have rich attribute information. For example, in the paper citation network, nodes represent papers, edges represent citation relationships, and the attribute information on nodes is the content of abstracts or full texts of papers. How to efficiently min...

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): G06K9/62G06N3/08
CPCG06N3/08G06F18/23213G06F18/217Y02D10/00
Inventor 王佳麟高军白金泽李朝张吉王佳
Owner PEKING UNIV
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