Robust Model Design by Comparative Evaluation of Clustering Algorithms
Open Access
- 17 August 2023
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 11, 88135-88151
Abstract
The K-means algorithm, widely used in cluster analysis, is a centroid-based clustering method known for its high efficiency and scalability. However, in realistic situations, the operating environment is susceptible to contamination issues caused by outliers and distribution departures, which may lead to clustering results from K-means that are distorted or rendered invalid. In this paper, we introduce three other alternative algorithms, including K-weighted-medians, K-weighted-L 2 -medians, and K-weighted-HLs, to address these issues under the consideration of data with weights. The impact of contamination is investigated by examining the estimation effects on optimal cluster centroids. We explore the robustness of the clustering algorithms from the perspective of the breakdown point, and then conduct experiments on simulated and real datasets to evaluate their performance using new numerical metrics. The results demonstrate the effectiveness of the proposed K-weighted-HLs algorithm, surpassing other algorithms in scenarios involving both contamination issues.Keywords
Funding Information
- National Research Foundation of Korea (NRF) grants
- Korean Government [Ministry of Science and Information and communications technology (MSIT)] (2022R1A2C1091319, RS-2023-00242528, 2021R1F1A1052846)
This publication has 45 references indexed in Scilit:
- OR models in urban service facility location: A critical review of applications and future developmentsEuropean Journal of Operational Research, 2019
- Big-Data Analysis, Cluster Analysis, and Machine-Learning Approaches2018
- Consumer perspectives about mobile banking adoption in India – a cluster analysisInternational Journal of Bank Marketing, 2017
- A highly efficient robust design under data contaminationComputers & Industrial Engineering, 2016
- Brain tumor segmentation based on a hybrid clustering techniqueEgyptian Informatics Journal, 2015
- Using K-Means Clustering and Parameter Weighting for Partial-Discharge Noise SuppressionIEEE Transactions on Power Delivery, 2011
- On Two Types of Breakdown Points of WeightedL2-medianCommunications in Statistics - Theory and Methods, 2011
- Fuzzy PCA-Guided Robust $k$-Means ClusteringIEEE Transactions on Fuzzy Systems, 2009
- Kernel density estimation and K-means clustering to profile road accident hotspotsAccident Analysis & Prevention, 2009
- Analysis and Implementation of an Adaptive Algorithm for the Rejection of Multiple Sinusoidal DisturbancesIEEE Transactions on Control Systems Technology, 2008