Cited By
View all- Zhou SLiu FSong W(2021)Estimating the Optimal Number of Clusters Via Internal Validity IndexNeural Processing Letters10.1007/s11063-021-10427-853:2(1013-1034)Online publication date: 1-Apr-2021
Self-Organizing-Map Based Clustering Using a Local Clustering Validity Index
Pages 253 - 271
Abstract
Classical clustering methods, such as partitioning and hierarchical clustering algorithms, often fail to deliver satisfactory results, given clusters of arbitrary shapes. Motivated by a clustering validity index based on inter-cluster and intra-cluster density, we propose that the clustering validity index be used not only globally to find optimal partitions of input data, but also locally to determine which two neighboring clusters are to be merged in a hierarchical clustering of Self-Organizing Map (SOM). A new two-level SOM-based clustering algorithm using the clustering validity index is also proposed. Experimental results on synthetic and real data sets demonstrate that the proposed clustering algorithm is able to cluster data in a better way than classical clustering algorithms on an SOM.
References
[1]
1. Pal, N. R., Bezdek, J. C. and Tsao, E. C.-K.: Generalized clustering networks and Kohonen's self-organizing scheme, IEEE Trans. Neural Networks, 4(4) (1993), 549-557.
[2]
2. Jain, A.K., Murty, M. N. and Flyn, P. J.: Data clustering: a review, ACM Computing Surveys, 31(3) (1999), 264-323.
[3]
3. Han, J. and Kamber, M.: Data Mining: Concepts and Techniques, Morgan-Kaufman, San Francisco, 2000.
[4]
4. Kohonen, T.: Self-Organizing Maps, Springer-Verlag, Berlin, Germany, 1997.
[5]
5. Huntsberger, T. and Ajjimarangsee, P.: Parallel self-organizing feature maps for unsupervised pattern recognition, Int. J. General systems, 16 (1989), 357-372.
[6]
6. Mao, J. and Jain, A. K.: A self-organizing network for hyperellipsoidal clustering (HEC), IEEE Trans. on Neural Networks, 7(1) (1996), 16-29.
[7]
7. Lampinen, J. and Oja, E.: Clustering properties of hierarchical self-organizing maps, J. Math. Imag. Vis., 2(2-3) (1992), 261-272.
[8]
8. Murtagh, F.: Interpreting the Kohonen self-organizing feature map using contiguity-constrained clustering, Pattern Recognition Letters, 16 (1995), 399-408.
[9]
9. Kiang, M. Y.: Extending the Kohonen self-organizing map networks for clustering analysis, Computational Statistics & Data Analysis, 38 (2001), 161-180.
[10]
10. Cottrell, M., Gaubert, P., Letremy P. and Rousset, P.: Analyzing and representing multi-dimensional quantitative and qualitative data: Demographic study of the Rhône valley. The domestic consumption of the Canadian families, In: Erkki Oja and Samuel Kaski (eds.), Kohonen maps, pp. 1-14, Elsevier, 1999.
[11]
11. Vesanto, J. and Alhonierni, E.: Clustering of the self-organizing map, IEEE Trans. Neural Networks, 11(3) (2000), 586-600.
[12]
12. Lopson, H. and Siegelmann, H. T.: Clustering irregular shapes using high-order neurons, Neural Computation, 12(10) (2000), 2331-2353.
[13]
13. Halkidi, M. and Vazirgiannis, M.: Clustering validity assessment using multi representatives, In: Proceedings of SETN Conference, Thessaloniki, Greece, April 2002.
[14]
14. Guha, S., Rastogi, R. and Shim, K.: CURE: An efficient clustering algorithm for large databases, In: Proceedings of ACM SIGMOD International Conference on Management of Data, New York, pp. 73-84, 1998.
[15]
15. Duda, O., Hart, P. E. and Stork, D. G.: Pattern Classification, John Wiley & Sons, USA, 2001.
[16]
16. Fisher, R. A.: The use of multiple measure in taxonomic problems, Ann. Eugenics, 7(Part II) (1936), 179-188.
[17]
17. Bezdek, J. C. and Pal, N. R.: Some new indexes of cluster validity, IEEE Trans. on System, Man, and Cybernetics, 28(3) (1998), 301-315.
[18]
18. Alon, U., Barkai, N., Notterman, D. A., Gish, G., Ybarra, S., Mack, D. and Levine, A. J.: Broad patterns of gene expression revealed by clustering analysis of tumor and normal colon tissues probed by olignucleotide arrays, Proc. Natl. Acad. Sci., 96 (1999), 6745-6750.
Index Terms
- Self-Organizing-Map Based Clustering Using a Local Clustering Validity Index
Recommendations
A new validity clustering index-based on finding new centroid positions using the mean of clustered data to determine the optimum number of clusters
Highlights- A new CVI called VCIM is proposed to validate the clustering algorithm results.
AbstractClustering, an unsupervised pattern classification method, plays an important role in identifying input dataset structures. It partitions input datasets into clusters or groups where either the optimum number of clusters is known in ...
An effective partitional clustering algorithm based on new clustering validity index
Highlights- The traditional K-means algorithm is improved by the density parameters based initial centers selection method.
AbstractAs an unsupervised pattern classification method, clustering partitions the input datasets into groups or clusters. It plays an important role in identifying the natural structure of the target datasets. Now, it has been widely used in ...
Fast and stable clustering analysis based on Grid-mapping K-means algorithm and new clustering validity index
Highlights- Proposes the Grid-K-means clustering algorithm.
- Designs a new clustering ...
AbstractAs a classical data analysis technique, clustering plays the important role in identifying natural structures of target datasets. However, many of the existing clustering methods, including clustering algorithms and clustering validity ...
Reviews
Fazli Can
Self-organizing mapping (SOM), or Kohonen mapping, a close cousin to k-means clustering, seeks to extract and visually display the topological structure of high-dimensional input data. In the case of clustering, similar objects are assigned to the same group, with no visualization attempt. In this study, a SOM-based clustering algorithm is proposed. For this purpose, the authors use a slightly modified version of a cluster validity index to determine which neighboring pairs of clusters are to be merged into one cluster. The experimental results are visually appealing, and, in terms of clustering accuracy, they are better than that of the other clustering algorithms used for comparison (k-means, and hierarchical clustering algorithms, such as single-link, complete-link, and average-link). For example, clustering accuracy with the widely used Iris data set is 96 percent using the proposed algorithm, while the accuracy values using the other algorithms are 85, 68, 84, and 69 percent, respectively. The experiments involved four data sets, with various numbers of data points and dimensions. Two of the data sets are synthetic. However, one should be careful when using synthetic data for clustering, since they may involve some hidden bias. The paper is easy to read, but its presentation includes flaws and needs improvement. For example, the paper includes statements (page 254) such as "the second SOM layer takes outputs of the first SOM layer as inputs of the second SOM layer," and typographical errors, such as the use of the word "wildly" instead of "widely" (page 264). These tricks of the English language may be difficult for nonnative speakers.
Access critical reviews of Computing literature here
Become a reviewer for Computing Reviews.
Comments
Information & Contributors
Information
Published In
Copyright © Copyright © 2003 Kluwer Academic Publishers.
Publisher
Kluwer Academic Publishers
United States
Publication History
Published: 01 June 2003
Author Tags
Qualifiers
- Article
Contributors
Other Metrics
Bibliometrics & Citations
Bibliometrics
Article Metrics
- View Citations1Total Citations
- 0Total Downloads
- Downloads (Last 12 months)0
- Downloads (Last 6 weeks)0
Reflects downloads up to 09 Jan 2025
Other Metrics
Citations
Cited By
View all- Zhou SLiu FSong W(2021)Estimating the Optimal Number of Clusters Via Internal Validity IndexNeural Processing Letters10.1007/s11063-021-10427-853:2(1013-1034)Online publication date: 1-Apr-2021