Learning about the Learning Process

  • João Gama
  • Petr Kosina
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7014)

Abstract

This work addresses the problem of mining data stream generated in dynamic environments where the distribution underlying the observations may change over time. We present a system that monitors the evolution of the learning process. The system is able to self-diagnosis degradations of this process, using change detection mechanisms, and self-repairs the decision models. The system uses meta-learning techniques that characterize the domain of applicability of previously learned models. The meta-learns can detect re-occurrence of contexts, using unlabeled examples, and take pro-active actions by activating previously learned models.

Keywords

Data streams concept drift meta-learning recurrent concepts 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Dijkstra, W.: Self-stabilizing systems in spite of distributed control. Communications of the ACM 17(11), 643–644 (1974)CrossRefMATHGoogle Scholar
  2. 2.
    Gama, J., Medas, P., Castillo, G., Rodrigues, P.: Learning with drift detection. In: Bazzan, A.L.C., Labidi, S. (eds.) SBIA 2004. LNCS (LNAI), vol. 3171, pp. 286–295. Springer, Heidelberg (2004)CrossRefGoogle Scholar
  3. 3.
    Granitzer, M., Kröll, M., Seifert, C., Rath, A.S., Weber, N., Dietzel, O., Lindstaedt, S.N.: Analysis of machine learning techniques for context extraction. In: Pichappan, P., Abraham, A. (eds.) ICDIM, pp. 233–240. IEEE, Los Alamitos (2008)Google Scholar
  4. 4.
    Grant, E., Leavenworth, R.: Statistical Quality Control. McGraw-Hill, New York (1996)MATHGoogle Scholar
  5. 5.
    Harries, M.B., Sammut, C., Horn, K.: Extracting hidden context. Machine Learning 32, 101–126 (1998)CrossRefMATHGoogle Scholar
  6. 6.
    Katakis, I., Tsoumakas, G., Vlahavas, I.: Tracking recurring contexts using ensemble classifiers: an application to email filtering. Knowledge and Information Systems 22, 371–391 (2010)CrossRefGoogle Scholar
  7. 7.
    Klinkenberg, R.: Learning drifting concepts: Example selection vs. example weighting. Intelligent Data Analysis 8(3), 281–300 (2004)Google Scholar
  8. 8.
    Lazarescu, M.M.: A multi-resolution learning approach to tracking concept drift and recurrent concepts. In: Proceedings of the 5th International Workshop on Pattern Recognition in Information Systems (2005)Google Scholar
  9. 9.
    Ortega, J.: Exploiting multiple existing models and learning algorithms. In: AAAI 1996 - Workshop in Induction of Multiple Learning Models, pp. 17–21 (1995)Google Scholar
  10. 10.
    Ortega, J., Koppel, M., Argamon, S.: Arbitrating among competing classifiers using learned referees. Knowledge and Information Systems 3(4), 470–490 (2001)CrossRefMATHGoogle Scholar
  11. 11.
    Ramamurthy, S., Bhatnagar, R.: Tracking recurrent concept drift in streaming data using ensemble classifiers. In: ICMLA 2007: Proceedings of the Sixth International Conference on Machine Learning and Applications, pp. 404–409. IEEE Computer Society, Washington, DC, USA (2007)Google Scholar
  12. 12.
    Seewald, A., Fürnkranz, J.: An evaluation of grading classifiers. In: Hoffmann, F., Hand, D.J., Adams, N., Fisher, D., Guimaraes, G. (eds.) IDA 2001. LNCS, vol. 2189, pp. 115–124. Springer, Heidelberg (2001)CrossRefGoogle Scholar
  13. 13.
    Nick Street, W., Kim, Y.: A streaming ensemble algorithm (sea) for large-scale classification. In: Knowledge Discovery and Data Mining, pp. 377–382. ACM Press, New York (2001)Google Scholar
  14. 14.
    Turney, P.: The management of context-sensitive features: A review of strategies (1996)Google Scholar
  15. 15.
    Widmer, G.: Tracking context changes through meta-learning. Machine Learning 27(3), 259–286 (1997)CrossRefGoogle Scholar
  16. 16.
    Widmer, G., Kubat, M.: Learning in the presence of concept drift and hidden contexts. Machine Learning 23(1), 69–101 (1996)Google Scholar
  17. 17.
    Yang, Y., Wu, X., Zhu, X.: Mining in anticipation for concept change: Proactive-reactive prediction in data streams. Data Mining and Knowledge Discovery 13(3), 261–289 (2006)MathSciNetCrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • João Gama
    • 1
  • Petr Kosina
    • 2
  1. 1.LIAAD-INESC PortoFEP-University of PortoPortugal
  2. 2.LIAAD-INESC PortoFI Masaryk UniversityCzech Republic

Personalised recommendations