Sign prediction method capable of considering local labelling characteristics of node

Abstract

The invention provides a sign prediction method capable of considering the local labelling characteristics of a node. The method comprises the following steps that: S1: obtaining the form expression of a signed network; S2: analyzing the structural characteristics of the signed network, obtaining a relationship between the rank and structure balance of a network adjacent matrix and a weak structure balance theory, utilizing low-rank matrix decomposition to decompose a low-rank matrix, and meeting a constraint condition that a matrix rank is minimum; S3: according to the conclusion of the structural characteristics of the signed network, converting a sign prediction problem into an optimization problem; and S4: utilizing a stochastic gradient descent method to decompose the low-rank matrix to obtain network considering global characteristics.

Description

technical field [0001] The invention relates to an Internet technology, in particular to a symbol prediction method considering the local labeling characteristics of nodes. Background technique [0002] A sign network refers to a network whose edges have positive or negative sign attributes. A positive sign indicates that there is a mutual trust and positive friendship between two users in the network, while a negative edge indicates a distrustful and negative hostile relationship. Networks with symbolic properties are ubiquitous because adversarial relationships exist in many complex systems: users in online social networks can express trust or distrust toward other users, mark friend or enemy relations, give different opinions on a certain issue In biological systems, there are facilitation and inhibition between neurons; in international relations, there are cooperation and hostility. Studying the symbolic properties of the network is conducive to understanding the basic ...

AI Technical Summary

Problems solved by technology

These complex systems cannot accurately reflect the relationship between nodes with the existing unsigned binary network modeling, and the research ideas and methods of simple networks are not suitable for symbolic networks.

Images