Addressing limitations of the K-means clustering algorithm: outliers, non-spherical data, and optimal cluster selection

Author:

khan Iliyas Karim1,Daud Hanita Binti1,Zainuddin Nooraini binti1,Sokkalingam Rajalingam1,Abdussamad 1,Museeb Abdul1,Inayat Agha2

Affiliation:

1. Fundamental and Applied Science Department, Universiti Teknologi PETRONAS, Perak 32610, Malaysia

2. Department of Statistic University of Malakand Chakdara Lower Dir, Khyber Pakhtunkhwa Pakistan

Abstract

<p>Clustering is essential in data analysis, with K-means clustering being widely used for its simplicity and efficiency. However, several challenges can affect its performance, including the handling of outliers, the transformation of non-spherical data into a spherical form, and the selection of the optimal number of clusters. This paper addressed these challenges by developing and enhancing specific models. The primary objective was to improve the robustness and accuracy of K-means clustering in the presence of these issues. To handle outliers, this research employed the winsorization method, which uses threshold values to minimize the influence of extreme data points. For the transformation of non-spherical data into a spherical form, the KROMD method was introduced, which combines Manhattan distance with a Gaussian kernel. This approach ensured a more accurate representation of the data, facilitating better clustering performance. The third objective focused on enhancing the gap statistic for selecting the optimal number of clusters. This was achieved by standardizing the expected value of reference data using an exponential distribution, providing a more reliable criterion for determining the appropriate number of clusters. Experimental results demonstrated that the winsorization method effectively handles outliers, leading to improved clustering stability. The KROMD method significantly enhanced the accuracy of converting non-spherical data into spherical form, achieving an accuracy level of 0.83 percent and an execution time of 0.14 per second. Furthermore, the enhanced gap statistic method outperformed other techniques in selecting the optimal number of clusters, achieving an accuracy of 93.35 percent and an execution time of 0.1433 per second. These advancements collectively enhance the performance of K-means clustering, making it more robust and effective for complex data analysis tasks.</p>

Publisher

American Institute of Mathematical Sciences (AIMS)

Reference49 articles.

1. X. Du, Y. He, J. Z. Huang, Random sample partition-based clustering ensemble algorithm for big data, 2021 IEEE International Conference on Big Data (Big Data), 2021, 5885–5887. https://doi.org/10.1109/BigData52589.2021.9671297

2. B. Huang, Z. Liu, J. Chen, A. Liu, Q. Liu, Q. He, Behavior pattern clustering in blockchain networks, Multimed. Tools Appl., 76 (2017), 20099–20110. https://doi.org/10.1007/s11042-017-4396-4

3. Y. Djenouri, A. Belhadi, D. Djenouri, J. C. W. Lin, Cluster-based information retrieval using pattern mining, Appl. Intell., 51 (2021), 1888–1903. https://doi.org/10.1007/s10489-020-01922-x

4. C. Ouyang, C. Liao, D. Zhu, Y. Zheng, C. Zhou, C. Zou, Compound improved Harris hawks optimization for global and engineering optimization, Cluster Comput., 2024. https://doi.org/10.1007/s10586-024-04348-z

5. J. Xu, T. Li, D. Zhang, J. Wu, Ensemble clustering via fusing global and local structure information, Expert Syst. Appl., 237 (2024), 121557. https://doi.org/10.1016/j.eswa.2023.121557

同舟云学术

1.学者识别学者识别

2.学术分析学术分析

3.人才评估人才评估

"同舟云学术"是以全球学者为主线,采集、加工和组织学术论文而形成的新型学术文献查询和分析系统,可以对全球学者进行文献检索和人才价值评估。用户可以通过关注某些学科领域的顶尖人物而持续追踪该领域的学科进展和研究前沿。经过近期的数据扩容,当前同舟云学术共收录了国内外主流学术期刊6万余种,收集的期刊论文及会议论文总量共计约1.5亿篇,并以每天添加12000余篇中外论文的速度递增。我们也可以为用户提供个性化、定制化的学者数据。欢迎来电咨询!咨询电话:010-8811{复制后删除}0370

www.globalauthorid.com

TOP

Copyright © 2019-2024 北京同舟云网络信息技术有限公司
京公网安备11010802033243号  京ICP备18003416号-3