A network representation method based on depth network structure and node attributes

A network structure and network representation technology, applied in the field of network analysis, can solve the problem of not being able to comprehensively consider high-order information between nodes and node attribute information, and achieve the effect of accurate representation and improved accuracy.

A network structure and network representation technology, applied in the field of network analysis, can solve the problem of not being able to comprehensively consider high-order information between nodes and node attribute information, and achieve the effect of accurate representation and improved accuracy.

CN109101629AInactive Publication Date: 2018-12-28HEFEI UNIV OF TECH
3 Cites 11 Cited by

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • A network representation method based on depth network structure and node attributes
  • A network representation method based on depth network structure and node attributes
  • A network representation method based on depth network structure and node attributes

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0058] see figure 1 , the network representation method based on the deep network structure and node attributes in this embodiment is performed in the following steps:

[0059] Step 1. Construct the adjacency matrix S and attribute relationship matrix B between nodes:

[0060] Denote an attribute network with G, G=(V,E,X), where:

[0061] V represents the network node set, V={v 1 ,v 2 ,...,v N}, with v i and v j represent node i and node j respectively, i=1,2,...,N, j=1,2,...,N, N is the total number of nodes;

[0062] E represents the link set in the network, and the element e in the link set E i,j represents the node v i with node v j The link relationship between e i,j ∈E, and i≠j;

[0063] X represents the attribute matrix, and the attribute matrix X is a F×N real number matrix, which contains attribute information of all nodes, X={x 1 ,x 2 ,...,x N}, x i for node v i The attribute vector of F dimension;

[0064] Step 1.1. Obtain the adjacency matrix S bet...

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 representation method based on a depth network structure and node attribute, which comprehensively considers the influence of the network structure and node attributeinformation on the node, and learns the node characteristic representation through a neural network. The method Includes steps: an adjacency matrix and attribute relation matrix between nodes is constructed; the probability transfer matrix and attribute probability transfer matrix of the structure between nodes are obtained; according to the structural probability transfer matrix and the attribute probability transfer matrix, the multi-order probability relation matrix is obtained by using the personalized random walk model; the global information matrix is obtained by combining the multi-order probability relation matrix with the attenuation function; the global information matrix is inputted into the automatic encoder, and the low-dimensional feature representation of the network node is obtained by training the automatic encoder. The invention solves the problem of data sparsity, encodes the global information of nodes in the network into a low-dimensional, dense vector space by constructing a depth neural network, and ensures the accurate representation of nodes in the network.

Description

technical field [0001] The invention relates to the field of network analysis, in particular to a network representation based on deep network structure and node attributes. Background technique [0002] Network representation learning aims to encode symbolic data into a low-dimensional, continuous, dense vector space through an unsupervised method. With the rapid development of online social applications and media, a large amount of data reflecting the network structure has been generated. Learning the low-dimensional network representation of the network has shown good efficiency and effect in different application fields, including link prediction, network Node classification, anomaly detection, recommendation, etc. [0003] The biggest characteristic of the data on the network is that there is a link relationship between the sample points, which shows that the sample points in the network are not completely independent. In 2015, Microsoft Research proposed the LINE alg...

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
28 Dec 2018
Publication
CN109101629A
IPC
G06F17/30; G06N3/02
CPC
G06N3/02
Inventors
洪日昌; 何媛