Advertisement

SRF: A Framework for the Study of Classifier Behavior under Training Set Mislabeling Noise

  • Katsiaryna Mirylenka
  • George Giannakopoulos
  • Themis Palpanas
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7301)

Abstract

Machine learning algorithms perform differently in settings with varying levels of training set mislabeling noise. Therefore, the choice of a good algorithm for a particular learning problem is crucial. In this paper, we introduce the “Sigmoid Rule” Framework focusing on the description of classifier behavior in noisy settings. The framework uses an existing model of the expected performance of learning algorithms as a sigmoid function of the signal-to-noise ratio in the training instances. We study the parameters of the above sigmoid function using five different classifiers, namely, Naive Bayes, kNN, SVM, a decision tree classifier, and a rule-based classifier. Our study leads to the definition of intuitive criteria based on the sigmoid parameters that can be used to compare the behavior of learning algorithms in the presence of varying levels of noise. Furthermore, we show that there exists a connection between these parameters and the characteristics of the underlying dataset, hinting at how the inherent properties of a dataset affect learning. The framework is applicable to concept drift scenaria, including modeling user behavior over time, and mining of noisy data series, as in sensor networks.

Keywords

classification classifier evaluation handling noise concept drift 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Ali, S., Smith, K.A.: On learning algorithm selection for classification. Applied Soft Computing 6(2), 119–138 (2006)CrossRefGoogle Scholar
  2. 2.
    Bradley, A.P.: The use of the area under the ROC curve in the evaluation of machine learning algorithms. Pattern Recognition 30(7), 1145–1159 (1997)CrossRefGoogle Scholar
  3. 3.
    Camastra, F., Vinciarelli, A.: Estimating the intrinsic dimension of data with a fractal-based method. IEEE Transactions on Pattern Analysis and Machine Intelligence 24(10), 1404–1407 (2002)CrossRefGoogle Scholar
  4. 4.
    Chevaleyre, Y., Zucker, J.-D.: Noise-tolerant rule induction from multi-instance data. In: De Raedt, L. (ed.) Proceedings of the ICML 2000 Workshop on Attribute-Value and Relational Learning: Crossing the Boundaries (2000)Google Scholar
  5. 5.
    Cohen, W.W.: Fast effective rule induction. In: ICML (1995)Google Scholar
  6. 6.
    de Sousa, E., Traina, A., Traina Jr., C., Faloutsos, C.: Evaluating the intrinsic dimension of evolving data streams. In: Proceedings of the 2006 ACM Symposium on Applied Computing, pp. 643–648. ACM (2006)Google Scholar
  7. 7.
    Frank, A., Asuncion, A.: UCI machine learning repository (2010)Google Scholar
  8. 8.
    Giannakopoulos, G., Palpanas, T.: Adaptivity in entity subscription services. In: ADAPTIVE (2009)Google Scholar
  9. 9.
    Giannakopoulos, G., Palpanas, T.: Content and type as orthogonal modeling features: a study on user interest awareness in entity subscription services. International Journal of Advances on Networks and Services 3(2) (2010)Google Scholar
  10. 10.
    Giannakopoulos, G., Palpanas, T.: The effect of history on modeling systems’ performance: The problem of the demanding lord. In: ICDM (2010)Google Scholar
  11. 11.
    Giraud-Carrier, C., Vilalta, R., Brazdil, P.: Introduction to the special issue on meta-learning. Machine Learning 54(3), 187–193 (2004)CrossRefGoogle Scholar
  12. 12.
    Hall, M., Frank, E., Holmes, G., Pfahringer, B., Reutemann, P., Witten, I.H.: The weka data mining software: an update. ACM SIGKDD Explorations Newsletter 11(1), 10–18 (2009)CrossRefGoogle Scholar
  13. 13.
    Han, J., Kamber, M.: Data mining: concepts and techniques. Morgan Kaufmann (2006)Google Scholar
  14. 14.
    Kalapanidas, E., Avouris, N., Craciun, M., Neagu, D.: Machine learning algorithms: a study on noise sensitivity. In: Proc. 1st Balcan Conference in Informatics, pp. 356–365 (2003)Google Scholar
  15. 15.
    Keerthi, S.S., Shevade, S.K., Bhattacharyya, C., Murthy, K.R.K.: Improvements to platt’s smo algorithm for svm classifier design. Neural Computation 13(3), 637–649 (2001)zbMATHCrossRefGoogle Scholar
  16. 16.
    Kuh, A., Petsche, T., Rivest, R.L.: Learning time-varying concepts. In: NIPS, pp. 183–189 (1990)Google Scholar
  17. 17.
    Li, Q., Li, T., Zhu, S., Kambhamettu, C.: Improving medical/biological data classification performance by wavelet preprocessing. In: Proceedings ICDM Conference (2002)Google Scholar
  18. 18.
    Pendrith, M., Sammut, C.: On reinforcement learning of control actions in noisy and non-markovian domains. Technical report, School of Computer Science and Engineering, The University of New South Wales, Sydney, Australia (1994)Google Scholar
  19. 19.
    Teytaud, O.: Learning with noise. Extension to regression. In: Proceedings of International Joint Conference on Neural Networks, IJCNN 2001, vol. 3, pp. 1787–1792. IEEE (2002)Google Scholar
  20. 20.
    Theodoridis, S., Koutroumbas, K.: Pattern Recognition. Academic Press (2003)Google Scholar
  21. 21.
    Wolpert, D.: The existence of a priori distinctions between learning algorithms. Neural Computation 8, 1391–1421 (1996)CrossRefGoogle Scholar
  22. 22.
    Wolpert, D.: The supervised learning no-free-lunch theorems. In: Proc. 6th Online World Conference on Soft Computing in Industrial Applications. Citeseer (2001)Google Scholar
  23. 23.
    Wolpert, D.H.: The lack of a priori distinctions between learning algorithms. Neural Computation 8, 1341–1390 (1996)CrossRefGoogle Scholar
  24. 24.
    Wu, X., Kumar, V., Quinlan, J.R., Ghosh, J., Yang, Q., Motoda, H., McLachlan, G.J., Ng, A.F.M., Liu, B., Yu, P.S., Zhou, Z.-H., Steinbach, M., Hand, D.J., Steinberg, D.: Top 10 algorithms in data mining. Knowl. Inf. Syst. 14(1), 1–37 (2008)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2012

Authors and Affiliations

  • Katsiaryna Mirylenka
    • 1
  • George Giannakopoulos
    • 2
  • Themis Palpanas
    • 1
  1. 1.University of TrentoItaly
  2. 2.Institute of Informatics and Telecommunications of NCSR DemokritosGreece

Personalised recommendations