Optimized K-Medoids a High-Efficiency Clustering Technique for Big Data Analysis

K-Medoids clustering is a widely utilized algorithm for partitioning data into clusters, particularly when robustness to noise and outliers is crucial. Despite its advantages, traditional K-Medoids suffers from high computational complexity, making it impractical for large-scale datasets. This paper proposes an enhanced clustering method, termed Magnified K-Medoids, which integrates advanced medoid selection strategies, outlier detection mechanisms, and an adaptive cluster determination approach. The proposed method improves efficiency, scalability, and clustering quality, particularly for complex datasets with high dimensionality. Performance evaluation using the Higgs Boson dataset demonstrates that the Magnified K-Medoids algorithm surpasses traditional K-Medoids in clustering accuracy, execution time, and computational efficiency.

Keywords: Magnified K-Medoids, Clustering, Medoid Selection, Outlier Detection, Large-Scale Data.