A biogeography-based optimization algorithm for data clustering

Data clustering is a pivotal technique in data mining, essential for organizing data into meaningful groups across diverse domains such as engineering, medicine, and biology. This study introduces a Biogeography-based Optimization (BBO) algorithm to optimize data partitioning by effectively navigating the solution space towards optimal cluster configurations. The algorithm leverages migration and mutation mechanisms inspired by natural biogeography to enhance clustering accuracy. The proposed method is evaluated using various datasets of different scales and complexities, and its performance is benchmarked against conventional clustering algorithms, including K-means, Genetic Algorithm (GA), Simulated Annealing (SA), Ant Colony Optimization (ACO), and Particle Swarm Optimization (PSO). Comprehensive comparative analyses demonstrate that BBO not only achieves superior clustering accuracy but also exhibits robustness in handling diverse data distributions, underscoring its potential as a valuable tool in data clustering applications.

کلیدواژه ها:

data clustering ، Biogeography-based Optimization ، K-means ، ACO ، PSO ، GA