Authors: Rony Setyawan, Geraldi Catur Pamuji
Abstract
This study presents a comprehensive comparative analysis of the k-Means and Mean Shift clustering algorithms, utilizing waste data collected from West Java Province’s final disposal site spanning 2016 to 2021, with the primary objective of evaluating their performance and applicability for waste management practices; the analysis encompasses several critical parameters, including the number of clusters generated, variable uniformity, evaluation metrics employed, divergence measures, and processing time efficiency, revealing that k-Means, which formed three clusters, excels in rapid processing and provides finer cluster division, while Mean Shift, yielding two clusters, offers nuanced insights into data patterns, leading to the recommendation that the choice between the two algorithms should be driven by specific project requirements and considerations such as urgency of waste management needs and the depth of understanding desired for effective decision-making, thereby offering a tailored approach to waste data organization that optimizes categorization and contributes to more efficient and sustainable waste disposal practices in the future. © School of Engineering, Taylor’s University.
Keywords: k-means clustering; Mean shift clustering; Waste data
This article can be accessed at https://www.scopus.com/pages/publications/85197577372

