The DOI matches its Crossref record; 70 of the 70 references it cites were found in a registry or an index.
How much can k-means be improved by using better initialization and repeats?
Open accessAbstract
In this paper, we study what are the most important factors that deteriorate the performance of the k-means algorithm, and how much this deterioration can be overcome either by using a better initialization technique, or by repeating (restarting) the algorithm. Our main finding is that when the clusters overlap, k-means can be significantly improved using these two tricks. Simple furthest point heuristic (Maxmin) reduces the number of erroneous clusters from 15% to 6%, on average, with our clustering benchmark. Repeating the algorithm 100 times reduces it further down to 1%. This accuracy is more than enough for most pattern recognition applications. However, when the data has well separated clusters, the performance of k-means depends completely on the goodness of the initialization. Therefore, if high clustering accuracy is needed, a better algorithm should be used instead.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
(1967) · Cited by 23,719
Hierarchical Grouping to Optimize an Objective Function
Ward, J. H. (1963) · Journal of the American Statistical Association · doi:10.1080/01621459.1963.10500845 · Cited by 20,598
Least squares quantization in PCM
Lloyd, S. (1982) · IEEE Transactions on Information Theory · doi:10.1109/tit.1982.1056489 · Cited by 15,836
Algorithm AS 136: A K-Means Clustering Algorithm
Hartigan, J. A., & Wong, M. A. (1979) · Applied Statistics · doi:10.2307/2346830 · Cited by 14,523
(1973) · CERN Document Server (European Organization for Nuclear Research) · Cited by 12,895
Finding groups in data: an introduction to cluster analysis
(1991) · Choice Reviews Online · doi:10.5860/choice.28-4558 · Cited by 10,474
Data clustering: 50 years beyond K-means
Jain, A. K. (2010) · Pattern Recognition Letters · doi:10.1016/j.patrec.2009.09.011 · Cited by 9,368
(2007) · doi:10.5555/1283383.1283494 · Cited by 7,364
An Algorithm for Vector Quantizer Design
Linde, Y., Buzo, A., & Gray, R. (1980) · IEEE Transactions on Communications · doi:10.1109/tcom.1980.1094577 · Cited by 7,201
Clustering by fast search and find of density peaks
Rodriguez, A., & Laio, A. (2014) · Science · doi:10.1126/science.1242072 · Cited by 4,942
Pattern Classification and Scene Analysis
Thompson, M., Duda, R. O., & Hart, P. E. (1974) · Leonardo · doi:10.2307/1573081 · Cited by 4,495
The global k-means clustering algorithm
Likas, A., Vlassis, N., & J. Verbeek, J. (2003) · Pattern Recognition · doi:10.1016/s0031-3203(02)00060-2 · Cited by 3,733
Pattern Recognition Principles
(2009) · Classification Methods for Remotely Sensed Data, Second Edition · doi:10.1201/9781420090741.ch2 · Cited by 3,173
A Comparison of Document Clustering Techniques
Steinbach, M. S., Karypis, G., & Kumar, V. (2000) · Cited by 2,532
Cluster analysis of multivariate data : efficiency versus interpretability of classifications
Forgy, E. W. (1965) · Biometrics · Cited by 2,093
Clustering to minimize the maximum intercluster distance
Gonzalez, T. F. (1985) · Theoretical Computer Science · doi:10.1016/0304-3975(85)90224-5 · Cited by 1,717
Krishna, K., & Narasimha Murty, M. (1999) · IEEE Transactions on Systems, Man and Cybernetics, Part B (Cybernetics) · doi:10.1109/3477.764879 · Cited by 1,717
A comparative study of efficient initialization methods for the k-means clustering algorithm
Celebi, M. E., Kingravi, H. A., & Vela, P. A. (2013) · Expert Systems with Applications · doi:10.1016/j.eswa.2012.07.021 · Cited by 1,214
Refining Initial Points for K-Means Clustering
Bradley, P. S., & Fayyad, U. M. (1998) · Cited by 1,031
A clustering technique for summarizing multivariate data
Ball, G. H., & Hall, D. J. (1967) · Behavioral Science · doi:10.1002/bs.3830120210 · Cited by 912
An empirical comparison of four initialization methods for the K-Means algorithm
Peña, J. M., Lozano, J. A., & Larrañaga, P. (1999) · Pattern Recognition Letters · doi:10.1016/s0167-8655(99)00069-0 · Cited by 866
Efficient k-nearest neighbor graph construction for generic similarity measures
Dong, W., Moses, C., & Li, K. (2011) · Proceedings of the 20th international conference on World wide web · doi:10.1145/1963405.1963487 · Cited by 673
Bahmani, B., Moseley, B., Vattani, A., Kumar, R., & Vassilvitskii, S. (2012) · Proceedings of the VLDB Endowment · doi:10.14778/2180912.2180915 · Cited by 628
K-means properties on six clustering benchmark datasets
Fränti, P., & Sieranoja, S. (2018) · Applied Intelligence · doi:10.1007/s10489-018-1238-7 · Cited by 518
Self-citationFast approximate spectral clustering
Yan, D., Huang, L., & Jordan, M. I. (2009) · Proceedings of the 15th ACM SIGKDD international conference on Knowledge discovery and data mining · doi:10.1145/1557019.1557118 · Cited by 474
A new initialization technique for generalized Lloyd iteration
Katsavounidis, I. X., Kuo, C. ‐. J., & Zhang, Z. (1994) · IEEE Signal Processing Letters · doi:10.1109/97.329844 · Cited by 372
The k-means clustering technique: General considerations and implementation in Mathematica
Morissette, L., & Chartier, S. (2013) · Tutorials in Quantitative Methods for Psychology · doi:10.20982/tqmp.09.1.p015 · Cited by 327
Chiang, M. M.-T., & Mirkin, B. (2010) · Journal of Classification · doi:10.1007/s00357-010-9049-5 · Cited by 319
Initializing K-means Batch Clustering: A Critical Evaluation of Several Techniques
Steinley, D., & Brusco, M. J. (2007) · Journal of Classification · doi:10.1007/s00357-007-0003-0 · Cited by 316
Fast Agglomerative Clustering Using a k-Nearest Neighbor Graph
Fränti, P., Virmajoki, O., & Hautamäki, V. (2006) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2006.227 · Cited by 310
Self-citationClustering Stability: An Overview
Ulrike, V. L. (2010) · Foundations and Trends® in Machine Learning · doi:10.1561/2200000008 · Cited by 301
Local Optima in K-Means Clustering: What You Don't Know May Hurt You.
Steinley, D. (2003) · Psychological Methods · doi:10.1037/1082-989x.8.3.294 · Cited by 269
A method for initialising the K-means clustering algorithm using kd-trees
Redmond, S. J., & Heneghan, C. (2007) · Pattern Recognition Letters · doi:10.1016/j.patrec.2007.01.001 · Cited by 261
An efficient approximation to the K-means clustering for massive data
Capó, M., Pérez, A., & Lozano, J. A. (2017) · Knowledge-Based Systems · doi:10.1016/j.knosys.2016.06.031 · Cited by 218
Two improved k-means algorithms
Yu, S.-S., Chu, S.-W., Wang, C.-M., Chan, Y., & Chang, T.-C. (2017) · Applied Soft Computing · doi:10.1016/j.asoc.2017.08.032 · Cited by 212
Randomized Dimensionality Reduction for $k$ -Means Clustering
Boutsidis, C., Zouzias, A., Mahoney, M. W., & Drineas, P. (2015) · IEEE Transactions on Information Theory · doi:10.1109/tit.2014.2375327 · Cited by 207
A new algorithm for initial cluster centers in k-means algorithm
Erisoglu, M., Calis, N., & Sakallioglu, S. (2011) · Pattern Recognition Letters · doi:10.1016/j.patrec.2011.07.011 · Cited by 193
A new initialization method for categorical data clustering
Cao, F., Liang, J., & Bai, L. (2009) · Expert Systems with Applications · doi:10.1016/j.eswa.2009.01.060 · Cited by 188
IBM SPSS Statistics 19 Guide to Data Analysis
Spss, & Norušis, M. J. (2011) · Medical Entomology and Zoology · Cited by 186
A fast mean-distance-ordered partial codebook search algorithm for image vector quantization
Ra, S.-W., & Kim, J.-K. (1993) · IEEE Transactions on Circuits and Systems II: Analog and Digital Signal Processing · doi:10.1109/82.257335 · Cited by 179
Set Matching Measures for External Cluster Validity
Rezaei, M., & Franti, P. (2016) · IEEE Transactions on Knowledge and Data Engineering · doi:10.1109/tkde.2016.2551240 · Cited by 178
Fast density clustering strategies based on the k-means algorithm
Bai, L., Cheng, X., Liang, J., Shen, H., & Guo, Y. (2017) · Pattern Recognition · doi:10.1016/j.patcog.2017.06.023 · Cited by 178
Robust level set image segmentation via a local correntropy-based K-means clustering
Wang, L., & Pan, C. (2013) · Pattern Recognition · doi:10.1016/j.patcog.2013.11.014 · Cited by 151
Centroid index: Cluster level similarity measure
Fränti, P., Rezaei, M., & Zhao, Q. (2014) · Pattern Recognition · doi:10.1016/j.patcog.2014.03.017 · Cited by 148
Self-citationIn search of deterministic methods for initializing K-means and Gaussian mixture clustering
Su, T., & Dy, J. (2007) · Intelligent Data Analysis · doi:10.3233/ida-2007-11402 · Cited by 146
Frandsen, P. B., Calcott, B., Mayer, C., & Lanfear, R. (2015) · BMC Evolutionary Biology · doi:10.1186/s12862-015-0283-7 · Cited by 145
Density-based multiscale data condensation
Mitra, P., Murthy, C. A., & Pal, S. K. (2002) · IEEE Transactions on Pattern Analysis and Machine Intelligence · doi:10.1109/tpami.2002.1008381 · Cited by 139
An initialization method for the K -Means algorithm using neighborhood model
Cao, F., Liang, J., & Jiang, G. (2009) · Computers & Mathematics with Applications · doi:10.1016/j.camwa.2009.04.017 · Cited by 133
Initialization of cluster refinement algorithms: a review and comparative study
He, J., Lan, M., Tan, C.-L., Sung, S.-Y., & Low, H.-B. (2005) · International Joint Conference on Neural Networks (IJCNN) · doi:10.1109/ijcnn.2004.1379917 · Cited by 127
The $K$-Means-Type Algorithms Versus Imbalanced Data Distributions
Liang, J., Bai, L., Dang, C., & Cao, F. (2012) · IEEE Transactions on Fuzzy Systems · doi:10.1109/tfuzz.2011.2182354 · Cited by 125
Randomised Local Search Algorithm for the Clustering Problem
Fränti, P., & Kivijärvi, J. (2000) · Pattern Analysis and Applications · doi:10.1007/s100440070007 · Cited by 113
Self-citationNew methods for the initialisation of clusters
Al-Daoud, M. B., & Roberts, S. (1996) · Pattern Recognition Letters · doi:10.1016/0167-8655(95)00119-0 · Cited by 96
Efficient and portable combined Tausworthe random number generators
Tezuka, S., & L’Ecuyer, P. (1991) · ACM Transactions on Modeling and Computer Simulation · doi:10.1145/116890.116892 · Cited by 93
A comparison of several vector quantization codebook generation approaches
Huang, C.-C., & Harris, R. W. (1993) · IEEE Transactions on Image Processing · doi:10.1109/83.210871 · Cited by 91
Genetic algorithm with deterministic crossover for vector quantization
Fränti, P. (2000) · Pattern Recognition Letters · doi:10.1016/s0167-8655(99)00133-6 · Cited by 78
Self-citationComparison of clustering methods: A case study of text-independent speaker modeling
Kinnunen, T., Sidoroff, I., Tuononen, M., & Fränti, P. (2011) · Pattern Recognition Letters · doi:10.1016/j.patrec.2011.06.023 · Cited by 73
Self-citationOn -means algorithm with the use of Mahalanobis distances
Melnykov, I. V., & Melnykov, V. (2013) · Statistics & Probability Letters · doi:10.1016/j.spl.2013.09.026 · Cited by 72
Fast and memory efficient implementation of the exact PNN
Fränti, P., Kaukoranta, T., Shen, D.-F., & Chang, K.-S. (2000) · IEEE Transactions on Image Processing · doi:10.1109/83.841516 · Cited by 68
Self-citationA fast exact GLA based on code vector activity detection
Kaukoranta, T., Fränti, P., & Nevalainen, O. (2000) · IEEE Transactions on Image Processing · doi:10.1109/83.855429 · Cited by 67
Self-citationA better tree-structured vector quantizer
Wu, X., & Zhang, K. (2002) · DCC · doi:10.1109/dcc.1991.213341 · Cited by 64
Deterministic Initialization of the K-Means Algorithm Using Hierarchical ClusteringPreprint
Celebi, M. E., & Kingravi, H. A. (2012) · Cited by 61
DETERMINISTIC INITIALIZATION OF THE K-MEANS ALGORITHM USING HIERARCHICAL CLUSTERING
CELEBI, M. E., & KINGRAVI, H. A. (2012) · International Journal of Pattern Recognition and Artificial Intelligence · doi:10.1142/s0218001412500188 · Cited by 56
Can the Number of Clusters Be Determined by External Indices?
Rezaei, M., & Fränti, P. (2020) · IEEE Access · doi:10.1109/access.2020.2993295 · Cited by 44
Self-citationDensity core-based clustering algorithm with dynamic scanning radius
Xie, J., Xiong, Z., Zhang, Y., Feng, Y., & Ma, J. (2017) · Knowledge-Based Systems · doi:10.1016/j.knosys.2017.11.025 · Cited by 42
On the splitting method for vector quantization codebook generation
Franti, P., Kaukoranta, T., & Nevalainen, O. (1997) · Optical Engineering · doi:10.1117/1.601531 · Cited by 41
Common Nearest Neighbor Clustering—A Benchmark
Lemke, O., & Keller, B. G. (2018) · Algorithms · doi:10.3390/a11020019 · Cited by 26
A novel and simple strategy for evolving prototype based clustering
Márquez, D. G., Otero, A., Félix, P., & Martínez, C. A. G. (2018) · Pattern Recognition · doi:10.1016/j.patcog.2018.04.020 · Cited by 25
Grouping points by shared subspaces for effective subspace clustering
Zhu, Y., Ting, K. M., & Carman, M. (2018) · Pattern Recognition · doi:10.1016/j.patcog.2018.05.027 · Cited by 18
Recent Developments in Model-Based Clustering with Applications
Melnykov, V., Michael, S., & Melnykov, I. V. (2014) · Springer eBooks · doi:10.1007/978-3-319-09259-1_1 · Cited by 13
Constructing a High-Dimensional k NN-Graph Using a Z-Order Curve
Sieranoja, S., & Fränti, P. (2018) · ACM Journal of Experimental Algorithmics · doi:10.1145/3274656 · Cited by 10
Self-citation
Cited by
Ikotun, A. M., Ezugwu, A. E., Abualigah, L., Abuhaija, B., & Heming, J. (2023) · Information Sciences · doi:10.1016/j.ins.2022.11.139 · Cited by 2,272
An, Q., Rahman, S., Zhou, J., & Kang, J. J. (2023) · Sensors · doi:10.3390/s23094178 · Cited by 340
Metadata
- Authors
- Fränti, P., & Sieranoja, S.
- Venue
- Pattern Recognition
- Publisher
- Elsevier BV
- Published
- 2019-09
- Volume
- 93
- Pages
- 95-112
- ISSN
- 0031-3203 (ISSN-L)
- OpenAlex
- W2940320374
- Cited by
- 451
- References
- 70
- FWCI
- 24.98
- Type
- journal article
- Language
- en
- Open access
- hybrid
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.