Density-based particle swarm optimization algorithm for data clustering
作者：
Highlights：
• Kernel density-based particle swarm optimization algorithm is proposed.
• Multi-dimensional gravitational learning factors of particles are introduced.
• Gaussian kernel is employed to find for the densest region in a cluster.
• New simple bandwidth estimation method of the kernel is presented.
• A framework balancing the exploration and exploitation processes is proposed.
摘要
•Kernel density-based particle swarm optimization algorithm is proposed.•Multi-dimensional gravitational learning factors of particles are introduced.•Gaussian kernel is employed to find for the densest region in a cluster.•New simple bandwidth estimation method of the kernel is presented.•A framework balancing the exploration and exploitation processes is proposed.
论文关键词：Particle swarm optimization,Swarm intelligence,Universal gravity rule,Kernel density estimation,Exploitation and exploration balance,Data clustering
论文评审过程：Received 18 May 2017, Revised 6 August 2017, Accepted 24 August 2017, Available online 1 September 2017, Version of Record 8 September 2017.
论文官网地址：https://doi.org/10.1016/j.eswa.2017.08.050
