Neighborhood-Based Smoothing of External Cluster Validity Measures

  • Ken-ichi Fukui
  • Masayuki Numao
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7301)


This paper proposes a methodology for introducing a neighborhood relation of clusters to the conventional cluster validity measures using external criteria, that is, class information. The extended measure evaluates the cluster validity together with connectivity of class distribution based on a neighborhood relation of clusters. A weighting function is introduced for smoothing the basic statistics to set-based measures and to pairwise-based measures. Our method can extend any cluster validity measure based on a set or pairwise of data points. In the experiment, we examined the neighbor component of the extended measure and revealed an appropriate neighborhood radius and some properties using synthetic and real-world data.


cluster validity neighborhood relation weighting function 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Amigó, E., Gonzalo, J., Artiles, J., Verdejo, F.: A comparison of extrinsic clustering evaluation metrics based on formal constraints. Information Retrieval 699(12), 461–486 (2009)CrossRefGoogle Scholar
  2. 2.
    Davies, D.L., Bouldin, D.W.: A cluster separation measure. IEEE Transactions on Pattern Analsis and Machine Intelligence (TPAMI) 1(4), 224–227 (1979)CrossRefGoogle Scholar
  3. 3.
    Deborah, L.J., Baskaran, R., Kannan, A.: A survey on internal validity measure for cluster validation. International Journal of Computer Science & Engineering Survey (IJCSES) 1(2), 85–102 (2010)CrossRefGoogle Scholar
  4. 4.
    Dunn, J.C.: Well separated clusters and optimal fuzzy partitions. Journal of Cybernetics 4, 95–104 (1974)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Halkidi, M., Vazirgiannis, M.: Clustering validity assessment using multi representatives. In: Proc. 2nd Hellenic Conference on Artificial Intelligence, pp. 237–248 (2002)Google Scholar
  6. 6.
    Kohonen, T.: Self-Organizing Maps. Springer (1995)Google Scholar
  7. 7.
    Kovács, F., Legány, C., Babos, A.: Cluster validity measurement techniques. Engineering 2006, 388–393 (2006)Google Scholar
  8. 8.
    Kremer, H., Kranen, P., Jansen, T., Seidl, T., Bifet, A., Holmes, G., Pfahringer, B.: An effective evaluation measure for clustering on evolving data streams. In: Proc. the 17th SIGKDD Conference on Knowledge Discovery and Data Mining (KDD 2011), pp. 868–876 (2011)Google Scholar
  9. 9.
    Liu, Y., Li, Z., Xiong, H., Gao, X., Wu, J.: Understanding of internal clustering validation measures. In: Proc. IEEE International Conference on Data Mining (ICDM 2010), pp. 911–916 (2010)Google Scholar
  10. 10.
    Rendón, E., Abundez, I., Arizmendi, A., Quiroz, E.M.: Internal versus external cluster validation indexes. International Journal of Computers and Communications 5(1), 27–34 (2011)Google Scholar
  11. 11.
    Tasdemir, K., Merényi, E.: A new cluster validity index for prototype based clustering algorithms based on inter- and intra-cluster density. In: Proc. International Joint Conference on Neural Networks (IJCNN 2007), pp. 2205–2211 (2007)Google Scholar
  12. 12.
    Veenhuis, C., Koppen, M.: Data Swarm Clustering, ch. 10, pp. 221–241. Springer (2006)Google Scholar
  13. 13.
    Weinberger, K.Q., Blitzer, J., Saul, L.K.: Distance metric learning for large margin nearest neighbor classification. Journal of Machine Learning Research (JMLR) 10, 207–244 (2009)zbMATHGoogle Scholar
  14. 14.
    Xu, R., Wunsch, D.: Cluster Validity. Computational Intelligence, ch. 10, pp. 263–278. IEEE Press (2008)Google Scholar
  15. 15.
    Zha, Z.J., Mei, T., Wang, M., Wang, Z., Hua, X.S.: Robust distance metric learning with auxiliary knowledge. In: Proc. International Joint Conference on Artificial Intelligence (IJCAI 2009), pp. 1327–1332 (2009)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Ken-ichi Fukui
    • 1
  • Masayuki Numao
    • 1
  1. 1.The Institute of Scientific and Industrial Research (ISIR)Osaka UniversityIbarakiJapan

Personalised recommendations