Method for detecting non-overlapping network communities and overlapping network communities based on two-stage strategy

Abstract

The invention discloses a method for detecting non-overlapping network communities and overlapping network communities based on a two-stage strategy, which solves the problem of incapability of multi-resolution analysis of an overlapping network community structure in the prior art. The method has the implementation steps of (1) loading network data; (2) initiating the population of network communities; (3) dividing the population of the network communities; (4) building a network community team; (5) detecting candidate network community division; (6) updating the population of network communities; (7) detecting local network communities; (8) updating the population of network communities; (9) judging whether iteration is terminated or not; (10) outputting non-overlapping network community division; (11) detecting nodes of overlapping network communities; (12) outputting an overlapping network community division result. A module density function and a membership function are introduced, so that multi-resolution analysis of the overlapping network community structure is realized, and the overlapping network communities with different overlapping degrees can be detected.

Description

technical field [0001] The invention belongs to the field of computer technology, and further relates to a community detection method based on non-overlapping and overlapping networks in the field of artificial intelligence. The invention can be used in practical applications such as social network, large-scale integrated circuit, biomacromolecular structure analysis and function prediction. Background technique [0002] Many complex systems in the real world can be represented as networks, such as the World Wide Web, power grids, biological networks, and social networks. In addition to network properties such as small-world effect and scale-free, another important property in complex networks is community structure. Community refers to a collection of nodes with high similarity or close connection in the network. Analyzing the community structure of the network in the real world has important guiding significance for production and life. In the community structure of the...

AI Technical Summary

Problems solved by technology

However, the disadvantage of this method is that it cannot detect non-overlapping network community structures, and the method converges slowly.

Images