Multi-objective rapid genetic method for community network detection

Abstract

The invention discloses a multi-objective rapid genetic method for community network detection. The method is used to search for community structures in complex community networks, and solves the problems of "premature" and low efficiency in the optimization process of traditional algorithms. The present invention transforms the problem of community division into a multi-objective optimization problem. Firstly, two objective functions of community score and community fitness are constructed, and an external elite gene bank is introduced to store non-inferior solutions with high fitness. For the external elite gene bank Existing duplicate individuals do not need to repeat decoding, calculate the fitness value of the individual and a series of processes, and then execute the genetic variation crossover operator to return a set of non-dominated solutions that are compromised between the two objective functions, and generate the adjacency of the graph after decoding Matrix, thus dividing a complex community network into multiple independent sub-networks. The simulation shows that the multi-objective fast genetic algorithm introduces the concept of external elite gene pool to greatly reduce the time complexity and improve the efficiency and speed of complex network detection.

Description

technical field [0001] The invention relates to a method for realizing multi-objective rapid inheritance of community network detection, and belongs to the technical field of communication. technical background [0002] At present, the selection of multi-objective solution sets and the efficiency of algorithms have always been a thorny problem encountered in practice. Due to the mutual constraints among multi-objectives, the optimization of one of the objectives must affect the performance of other objectives, so multi-objectives are not There is a definite solution, but a set of optimal solutions, called Pareto optimal solution. The traditional solution transforms the multi-objective problem into a single-objective problem through weighted summation. However, due to the determination of the weight value, researchers need to There are certain prior judgments for each target, so it is difficult for such a method to really solve the multi-target problem. And the present inven...

AI Technical Summary

Problems solved by technology

[0003] The purpose of the present invention is to solve the problem of community network clustering in complex networks. The method solves the optimal solution selection problem and optimization efficiency problem in the Pareto non-inferior solution through the fast genetic method, and realizes the multi-objective network well. The clustered Pareto solution set screened out the optimal solution

Images