Complex network community discovery method under adaptive evolutionary bat algorithm for self-media network data

Abstract

The present invention provides a method for discovering complex network communities under the self-adaptive evolutionary bat algorithm for self-media network data, comprising the following steps: S1, obtaining massive data, constructing a network structure model, and using the bat algorithm as an adaptation of the modularity function degree function, using a character-based encoding method, using the label propagation method to improve the initialization population; S2, transforming the individual speed of the bat algorithm into a mutation probability value, using the crossover operator and the mutation operator to calculate the position update, so as to realize the common bat algorithm. Adaptive evolution, the adaptive evolution bat algorithm is used to divide the network to obtain more accurate network community division results. Compared with other intelligent algorithms for community discovery, this algorithm has the advantages of fast convergence speed and high solution accuracy, and is more suitable for community discovery under large-scale networks.

Description

technical field [0001] The invention relates to the field of big data mining, in particular to a method for discovering complex network communities under the self-adaptive evolutionary bat algorithm for self-media network data. Background technique [0002] Complex systems can be modeled as complex networks for analysis, and communities can be viewed as subsystems with certain properties or functions in the system. Community structures exist in many complex networks, and the study of community structures in complex networks has important theoretical significance. First of all, through the community structure, the common characteristics and attributes of the node group can be mined, which is helpful to analyze the relationship between the whole and the part; secondly, based on the existing network information, the functions of similar nodes can be predicted, and the potential relationship between nodes Connection possibility; finally, the impact of community structure on dyn...

AI Technical Summary

Problems solved by technology

The existing technology uses a genetic algorithm based on binary matrix coding. Although this coding is convenient for interleaving operations, the coding and decoding are complicated, and the algorithm time complexity is O(n 3 ), and coding correction is required; the immune cloning operator is added to the discrete differential evolution algorithm, which improves the local development capability of the algorithm and increases the robustness of the algorithm; the prior art proposes a discrete particle swarm optimization algorithm, which can update the speed Redefining and applying community discovery in symbolic networks

Modularity functions suffer from a resolution-limited problem, i.e. such methods do not perform well in the presence of small communities in large-scale networks

Images