Inertia-Based Indices to Determine the Number of Clusters in K-Means: An Experimental Evaluation

被引：4

作者：

Rykov, Andrei ^{[1
]}

de Amorim, Renato Cordeiro ^{[2
]}

Makarenkov, Vladimir ^{[3
,4
]}

Mirkin, Boris ^{[1
,5
]}

机构：

[1] Natl Res Univ Higher Sch Econ, Dept Data Anal & Machine Intelligence, Moscow 101000, Russia

[2] Univ Essex, Comp Sci & Elect Engn Dept, Wivenhoe CO4 3SQ, England

[3] Imagia Cybernet, Montreal, PQ H3C 3P8, Canada

[4] Mila Quebec AI Inst, Montreal, PQ H2S 3H1, Canada

[5] Univ London, Dept Comp Sci & Informat Syst, London WC1E 7HX, England

来源：

IEEE ACCESS | 2024年 / 12卷

基金：

加拿大自然科学与工程研究理事会;

关键词：

Indexes; Clustering algorithms; Euclidean distance; Amplitude modulation; Partitioning algorithms; Computer science; K-means; number of clusters; inertia; elbow method; Calinski-Harabasz index; Hartigan rule; ALGORITHM;

D O I：

10.1109/ACCESS.2024.3350791

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper gives an experimentally supported review and comparison of several indices based on the conventional K-means inertia criterion for determining the number of clusters, K, in datasets, using the popular Silhouette width index as a benchmark. Our experiments involve a novel version of the Elbow index, defined using values of K two or three steps apart. We also discuss alternative ways of computing the inertia and summarizing its values. Even though there are no overall winners in our experiments, some of our results are very conclusive and can be used as a guide for indices determining the number of clusters in K-means.

引用

页码：11761 / 11773

页数：13

共 50 条

[41] Apache Mahout's k-Means vs. Fuzzy k-Means Performance Evaluation
Xhafa, Fatos
Bogza, Adriana
Caballe, Santi
Barolli, Leonard
[J]. 2016 8TH INTERNATIONAL CONFERENCE ON INTELLIGENT NETWORKING AND COLLABORATIVE SYSTEMS (INCOS), 2016, : 110 - 116
[42] Classification of countries based on development indices by using K-means and grey relational analysis
Basel, Sayel
Gopakumar, K. U.
Rao, R. Prabhakara
[J]. GEOJOURNAL, 2022, 87 (05) : 3915 - 3933
[43] A GENERALIZED k-MEANS PROBLEM FOR CLUSTERING AND AN ADMM-BASED k-MEANS ALGORITHM
Ling, Liyun
Gu, Yan
Zhang, Su
Wen, Jie
[J]. JOURNAL OF INDUSTRIAL AND MANAGEMENT OPTIMIZATION, 2024, 20 (06) : 2089 - 2115
[44] Automatic detection of outliers and the number of clusters in k-means clustering via Chebyshev-type inequalities
Olukanmi, Peter
Nelwamondo, Fulufhelo
Marwala, Tshilidzi
Twala, Bhekisipho
[J]. NEURAL COMPUTING & APPLICATIONS, 2022, 34 (08): : 5939 - 5958
[45] Classification of countries based on development indices by using K-means and grey relational analysis
Sayel Basel
K. U. Gopakumar
R. Prabhakara Rao
[J]. GeoJournal, 2022, 87 : 3915 - 3933
[46] Kernel Penalized K-means: A feature selection method based on Kernel K-means
Maldonado, Sebastian
Carrizosa, Emilio
Weber, Richard
[J]. INFORMATION SCIENCES, 2015, 322 : 150 - 160
[47] Automatic detection of outliers and the number of clusters in k-means clustering via Chebyshev-type inequalities
Peter Olukanmi
Fulufhelo Nelwamondo
Tshilidzi Marwala
Bhekisipho Twala
[J]. Neural Computing and Applications, 2022, 34 : 5939 - 5958
[48] K-means Split Revisited: Well-grounded Approach and Experimental Evaluation
Grigorev, Valentin
Chernishev, George
[J]. SIGMOD'16: PROCEEDINGS OF THE 2016 INTERNATIONAL CONFERENCE ON MANAGEMENT OF DATA, 2016, : 2251 - 2252
[49] Determination of the number of clusters used in fuzzy inference systems by means of K-means and modeling of dam volume: Kestel dam example
Kucukerdem, Tulay Sugra
Kilit, Murat
Saplioglu, Kemal
[J]. PAMUKKALE UNIVERSITY JOURNAL OF ENGINEERING SCIENCES-PAMUKKALE UNIVERSITESI MUHENDISLIK BILIMLERI DERGISI, 2019, 25 (08): : 962 - 967
[50] Technical Note: Using k-means clustering to determine the number and position of isocenters in MLC-based multiple target intracranial radiosurgery
Yock, Adam D.
Kim, Gwe-Ya
[J]. JOURNAL OF APPLIED CLINICAL MEDICAL PHYSICS, 2017, 18 (05): : 351 - 357

← 1 2 3 4 5 →