Clustering Algorithm Recommendation: A Meta-learning Approach

  • Daniel G. Ferrari
  • Leandro Nunes de Castro
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7677)


Meta-learning is a technique that aims at understanding what types of algorithms solve what kinds of problems. Clustering, by contrast, divides a dataset into groups based on the objects’ similarities without the need of previous knowledge about the objects’ labels. The present paper proposes the use of meta-learning to recommend clustering algorithms based on the feature extraction of unlabelled objects. The features of the clustering problems will be evaluated along with the ranking of different algorithms so that the meta-learning system can recommend accurately the best algorithms for a new problem.


clustering algorithm recommendation ranking meta-learning 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Xu, R., Wunsch, D.: Survey of clustering algorithms. IEEE Transactions on Neural Networks, 645–678 (2005)Google Scholar
  2. 2.
    Fayyad, U., Piatetsky-Shapiro, G., Smyth, P.: From data mining to knowledge discovery: an overview. In: Fayyad, U., Piatetsky-Shapiro, G., Smyth, P., Uthurusamy, R. (eds.) Advances in Knowledge Discovery and Data Mining. American Association for Artificial Intelligence (1996)Google Scholar
  3. 3.
    Brazdil, P., Giraud-Carrier, C., Soares, C., Vilalta, R.: Metalearing - Applications to Data Mining. Springer (2009)Google Scholar
  4. 4.
    Giraud-Carrier, C., Vilalta, R., Brazdil, P.: Introduction to the Special Issue on Meta-Learning. Machine Learning 54(3), 187–193 (2004)CrossRefGoogle Scholar
  5. 5.
    Michie, D., Spiegelhalter, D., Taylor, C., Campbell, J. (eds.): Machine Learning, Neural and Statistical Classification, Ellis Horwood, Upper Saddle River, USA (1994)Google Scholar
  6. 6.
    de Souto, M., Prudêncio, R., Soares, R., de Araujo, D., Costa, I., Ludermir, T., Shliep, A.: Ranking and Selecting Clustering Algorithms Using a Meta-Learning Approach. In: IEEE International Joint Conference on Neural Networks, Hong Kong, pp. 3729–3735 (2008)Google Scholar
  7. 7.
    Nascimento, A.C.A., Prudêncio, R.B.C., de Souto, M.C.P., Costa, I.G.: Mining Rules for the Automatic Selection Process of Clustering Methods Applied to Cancer Gene Expression Data. In: Alippi, C., Polycarpou, M., Panayiotou, C., Ellinas, G. (eds.) ICANN 2009, Part II. LNCS, vol. 5769, pp. 20–29. Springer, Heidelberg (2009)CrossRefGoogle Scholar
  8. 8.
    Aha, D.: Generalizing from case studies: A case study. In: Proceedings of the 9th Internacional Workshop on Machine Learning, Aberdeen, pp. 1–10 (1992)Google Scholar
  9. 9.
    Brazdil, P., Soares, C., Da Costa, J.: Ranking learning algorithms: Using IBL and meta-learning on accuracy and time results. Machine Learning 50, 251–277 (2003)zbMATHCrossRefGoogle Scholar
  10. 10.
    Frank, A., Asuncion, A.: UCI Machine Learning Repository, (accessed 2010)
  11. 11.
    Kalousis, A.: Algorithm Selection via Meta-Learning. University of Geneva, Genebra (2002)Google Scholar
  12. 12.
    de Souza, B.: Meta-aprendizagem aplicada à classificação de dados de expressão gênica. USP, São Carlos (2010)Google Scholar
  13. 13.
    Amigó, E., Gonzalo, J., Artiles, J., Verdejo, F.: A comparison of extrinsic clustering evaluation metrics based on formal constraints. Information Retrieval 12(5), 461–486 (2009)CrossRefGoogle Scholar
  14. 14.
    Spearman, C.: The Proof and Measurement of Association Between Two Things. American Journal of Psychology, 72–101 (1904)Google Scholar
  15. 15.
    Lilliefors, H.: On the Kolmogorov-Smirnov Test for Normality with Mean and Variance Unknown. Journal of the American Statistical Association 62(318), 399–402 (1967)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Daniel G. Ferrari
    • 1
  • Leandro Nunes de Castro
    • 1
  1. 1.Natural Computing Laboratory (LCoN)Mackenzie Presbyterian UniversitySão PauloBrazil

Personalised recommendations