Network community detecting method based on multi-objective memetic computation

Abstract

The invention discloses a network community detecting method based on multi-objective memetic computation, which mainly solves the problems that the traditional method is not high in resolution ratio, so local optimum is easily caused, further only a single division result is obtained, a hierarchical structure of a network cannot be obtained, and the like. The method has the realization steps: (1) establishing an adjacency matrix of a to-be-detected network; (2) initiating a network population; (3) generating a new individual; (4) updating the network population; (5) locally searching the network population; (6) judging whether cyclic algebra is reached or not; (7) calculating the modularity value of each individual in the network population; (8) detecting communities obtained after network division is carried out. The network community detecting method has the beneficial effects that the network population is initiated by adopting a labeling method and combining a multi-objective evolutionary algorithm and a stimulated annealing algorithm based on discomposition, the initial detection precision of the network is improved, the convergence of the algorithm is accelerated, the local optimization capability of the algorithm is improved, the local optimum is avoided, the resolution ratio of the algorithm is improved, and the hierarchical structure of the network can be found.

Description

technical field [0001] The invention belongs to the field of network technology, and further relates to a network community detection method based on multi-objective cipher calculation in the field of data mining. The invention can be used for network community structure detection and analysis. Background technique [0002] Networks are an effective form of representing the interrelationships among objects in many systems in the real world. Such as collaborative network, World Wide Web, biological network, communication network, transportation network, social network, etc., these systems can be represented by complex networks. Studying the community structure of complex networks not only has very important theoretical significance for analyzing network topology, understanding network functions, discovering hidden laws in networks, and predicting network behavior, but also has broad application prospects. It has been applied to metabolic networks Analysis, protein interactio...

AI Technical Summary

Problems solved by technology

The disadvantage of this method is that the detection accuracy is not high, and it is easy to fall into problems such as local optimum.

Images