An Analysis of K-Means, Particle Swarm Optimization and Genetic Algorithm with Data Clustering Technique.

Data clustering is a popular data analysis approach for organizing similar objects without needing to be monitored - a collection of data, into meaningful groups (or clusters). Many approaches from different disciplines have been proposed as solutions for clustering problems and each algorithm has its advantages and drawbacks. This research paper offered a brief overview and performance comparison analysis between different existing data mining clustering strategies based on Particle Swarm Optimization, Genetic Algorithm, K-means and Hybrid PSO Solutions. The reasoning behind choosing these algorithms is that PSO has, due to the applicability of its advanced hybrid variants, reached a remarkable position in this field, and the Genetic Algorithm based unsupervised clustering technique provides a stable clustering performance with less computational time required. However, we found that Hybrid PSO solutions have been shown to produce excellent results in a wide variety of real-world data. In this paper, a brief review of these algorithms applied to data clustering has been described.