A Community Discovery-Oriented Network Weak Node Search Method

Abstract

The invention discloses a network weak node search method oriented to community discovery, including: (1) calculating the shortest distance between each node in the input community network and the boundary degree of each edge; (2) creating a population; setting an attack scheme, expressing the attack scheme as changing the gene on the chromosome, that is, changing the edge label corresponding to the gene; and initializing the parameters of the genetic algorithm; (3) calculating the fitness of each individual in the population according to the predefined fitness evaluation method, according to According to the fitness of each individual, roulette selection is performed on the population to generate the next generation of individuals; (4) After the chromosomes corresponding to the individuals are crossed, the chromosomes are mutated according to the probability of changing sides corresponding to the genes on the chromosomes after crossing; (5) Steps (3) and (4) are executed iteratively until the preset maximum number of iterations is met, and the optimal individual is output as a network weak node.

Description

technical field [0001] The invention belongs to the field of community networks, and in particular relates to a community discovery-oriented network weak node search method. Background technique [0002] There are many networks with a community structure in real life, and the community network is one of them. Each node in the community network represents a person, and the connection edge represents the exchange of information or friendship between people. According to the density of the connection edge, the Everyone is divided into different clusters, and each cluster is a community. [0003] A community is not just a collection of nodes that are closely connected to each other, they will show similar characteristics in their field, therefore, analyzing the community structure in a social network through a community discovery algorithm can help people analyze the interaction between groups of people in the network, Information security, personalized recommendation and other...

AI Technical Summary

Problems solved by technology

The gradient descent method has the characteristics of simplicity, high accuracy, etc., but it is not suitable for discontinuous or nonlinear situations. Simulated annealing algorithm and genetic algorithm perform well in local search and global search respectively, but their parameter sensitivity is relatively low. Strong, the particle swarm optimization algorithm has fast search speed and simple algorithm, but it is not suitable for dealing with discrete optimization problems and is easy to fall into local optimum

Images