A Study for Comparing Various Meta-Heuristic Algorithms in Data Clustering on applying K-Means

  • Unique Paper ID: 163177
  • Volume: 10
  • Issue: 11
  • PageNo: 970-975
  • Abstract:
  • Clustering, a technique used across various fields like pattern recognition and healthcare, involves grouping data points based on their similarities. However, traditional clustering methods sometimes struggle with issues like sensitivity to initial conditions and the tendency to get stuck in suboptimal solutions. To address these challenges, researchers have turned to Nature-Inspired Algorithms (NIAs), which mimic the behaviour of natural systems to find optimal solutions efficiently. Three popular NIAs—Ant Colony Optimization (ACO), Particle Swarm Optimization (PSO), and Genetic Algorithms (GA)—have gained attention for their ability to tackle clustering problems effectively. These algorithms emulate the behaviour of ants, particles in a swarm, and genetic evolution, respectively, to search for high-quality solutions in a complex search space. In this paper, we aim to determine which NIA performs best for clustering tasks, considering factors such as convergence speed, computational efficiency, and solution quality. By conducting a comprehensive comparative study, we evaluate the performance of each algorithm and assess their effectiveness in finding optimal sets of clusters. Our findings indicate that the Genetic Algorithm (GA) stands out as the most promising approach, outperforming other NIAs in terms of convergence speed, execution time, and success rate in finding high-quality clusters. We support our conclusions with rigorous statistical tests and detailed analyses, providing strong evidence for the superiority of the GA method in clustering applications. Overall, this paper contributes to the understanding of NIAs in clustering and highlights the practical advantages of using GA for solving clustering problems effectively and efficiently.

Cite This Article

  • ISSN: 2349-6002
  • Volume: 10
  • Issue: 11
  • PageNo: 970-975

A Study for Comparing Various Meta-Heuristic Algorithms in Data Clustering on applying K-Means

Related Articles