Visualization method, system and device for multi-dimensional network node classification, and storage medium

Abstract

Embodiments of the invention disclose a visualization method, system and device for multi-dimensional network node classification, and a storage medium. The method is based on a network embedding technology of machine learning, the network embedding technology of machine learning is combined with a regularization mechanism and an attention mechanism to obtain a low-dimensional dense vector of each node in a multi-dimensional graph network, and the low-dimensional dense vectors form a low-dimensional embedded matrix. Based on a non-linear dimension reduction algorithm, the low-dimensional embedded matrix is projected to obtain a coordinate value of each node in multi-dimensional graph data in a two-dimensional space, and a classification result is presented by adopting label information of the nodes for color mapping and employing a visualization technology. According to the invention, low-dimensional embedding obtained in the embodiments of the invention fuses node close distance, node long distance and the attribute information of the node at the same time. The obtained low-dimensional embedded matrix is projected into a two-dimensional layout space based on the nonlinear dimension reduction algorithm, and the influence of various feature information in the original multi-dimensional graph network on node classification is visually displayed from a visual angle by adopting the visualization technology.

Description

technical field [0001] The present application relates to the field of network data processing, in particular to a visualization method, system, device and storage medium for classifying multi-dimensional network nodes. Background technique [0002] With the rapid development of science and technology in the past few decades, especially emerging technologies represented by the Internet and big data, which have penetrated into all aspects of life, human beings have entered an information age. Among them, the development of information technology represented by the Internet has made systems in the real world interact with each other in different ways, and their connections are getting closer. For a given system, its internal connection mode can be described by a network: each component in the system is abstracted as a vertex (or node), and the connection between components is abstracted as an edge; such as economic network, social network, biological network , transportation ...

AI Technical Summary

Problems solved by technology

For example, taking an online social network as an example, the same group of users has interactive relationships in Sina Weibo, WeChat, and QQ social accounts. If the network with multiple interactive relationships is expressed as a multi-relationship fusion network, it will not be clear Express the respective structural characteristics of the graph network in the same dimension, as well as the coupling information and interaction information between different dimensions

[0004] The existing technology has the following technical problems: 1) The classic network embedding technology based on random walk can capture the long-distance node neighbor information of the target node (that is, the global structural characteristics), but this method can only capture the structural characteristics of the graph network, Unable to capture the attribute characteristics of nodes; 2) Graph convolutional network technology can naturally fuse the attribute characteristics of nodes, but cannot capture the long-distance neighbor information of target nodes; 3) Network embedding technology and graph convolution network technology based on random walk Designed for single-layer homogeneous graph networks, it cannot be directly used in multi-dimensional graph networks; 4) The classic graph visualization technology based on nonlinear dimensionality reduction only describes the similarity characteristics represented by the node structure, without considering the graph network The influence of other features in the node similarity

Images