.png){kind=icon}
Return
Core node knowledge based multi-objective particle swarm optimization for dynamic community detection
.png){kind=icon}
PRE
AI
DOI：10.1016/j.cie.2022.108843.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Community detection aims to classify nodes with similar attributes, and it becomes more difficult when the network structure changes over time. In order to tap into the evolution of dynamic networks, nodes with high criticality in dynamic networks clearly need to be attended. In this paper, a novel dynamic community detection algorithm based on particle swarm optimization is proposed. The core nodes in the network are obtained by calculating the resistance distance of each node. The nodes associated with these core nodes that make up the constant community. The constant community should be retained is the knowledge gained from the evolution of the core nodes in two consecutive time steps. To ensure the quality of community division at the current time step and the smoothness of two consecutive time steps, a new particle updating parameter is proposed, which is based on the optimal individual at the previous time step and the population at the current time step. Compared with the other well-known related algorithms, experimental result on many real-world and synthetic networks have illustrated the higher accuracy and stability of the proposed.
Keywords:
Dynamic networks
Particle swarm optimization
Constant community
Community detection
Journal
IF:
6.5
Papers:
1.0W
Citations:
3.8W
Organization
Cited Papers
.png){kind=icon}
ShareSaveThe biosynthesis of astaxanthin. VI. the conversion of [14c]lutein and [14c] β-carotene in goldfish
PREAI
ShareSave
ShareSaveShareSaveShareSave
View more