Skip to main content

Table 3 PSO based clustering methods

From: SDPSO: Spark Distributed PSO-based approach for feature selection and cancer disease prognosis

Authors Title Contributions
Ghorpade-Aher et al. [25] Clustering Multidimensional Data with PSO based Algorithm An advanced PSO algorithm entitled as Subtractive Clustering based Boundary Restricted Adaptive Particle Swarm Optimization (SC-BR-APSO) algorithm for clustering multidimensional data. The authors compare their algorithm with several algorithms using nine different datasets and affirm results with a minimum error rate and a maximum convergence rate
Niknam et al. [26] An efficient hybrid evolutionary optimization algorithm based on PSO and SA for clustering A hybrid evolutionary programming based clustering algorithm, called PSO-SA, combining PSO and Simulated Annealing (SA), behind which the basic idea is to search around the global solution using SA and to increase the information exchange among particles using a mutation operator to escape local optima. The authors test their approach on three datasets, Iris, Wisconsin Breast Cancer, and Ripley’s Glass, and provide evidence of the effectiveness of PSO-SA in finding optimal clusters
Dudeja [27] Fuzzy-based modified particle swarm optimization algorithm for shortest path problems A method that reduces the cost and time consumption with the help of fuzzy rules. The author proposes an enhancement of the execution of Modified Particle Swarm Optimization (MPSO) to assess the most limited way calculation with fuzzy rules. This hybrid method entitled Fuzzy-based Modified Particle Swarm Optimization showed an improved encoding efficiency, time consumption and cost
Cai et al. [28] A Novel Clustering Algorithm Based on DPC and PSO A clustering algorithm based on Density Peaks Clustering (DPC) and PSO (PDPC) that aims to overcome the numerous disadvantages of the DPC algorithm such as its inability to automatically determine the cluster centers and the possibility of the selected cluster centers to fall into a local optimum, which is surmounted through the use of the PSO algorithm best known for its capacity to rapidly reach the cluster cente
Mahesa et al. [29] Optimization of fuzzy c-means clustering using particle swarm optimization in brain tumor image segmentation A clustering technique using the fuzzy c-means optimized through the use of the PSO algorithm labeled as (FCM-PSO). The study aims to prove that this optimization shows better results than the non optimized version of the fuzzy c-means. To do so, the authors tested with six brain tumor images and demonstrate that the use of the PSO to enhances the clustering results