Advertisement

Taking Advantage of Class-Specific Feature Selection

  • Bárbara B. Pineda-Bautista
  • Jesús Ariel Carrasco-Ochoa
  • José Fco. Martínez-Trinidad
Part of the Lecture Notes in Computer Science book series (LNCS, volume 5788)

Abstract

In this work, a new method for class-specific feature selection, which selects a possible different feature subset for each class of a supervised classification problem, is proposed. Since conventional classifiers do not allow using a different feature subset for each class, the use of a classifier ensemble and a new decision rule for classifying new instances are also proposed. Experimental results over different databases show that, using the proposed method, better accuracies than using traditional feature selection methods, are achieved.

Keywords

Class-specific feature selection Feature selection supervised classification classifier ensemble 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Dash, M., Liu, M.: Feature Selection for Classification. Intelligent Data Analysis 1, 131–156 (1997)CrossRefGoogle Scholar
  2. 2.
    John, G.H., Kohavi, R., Pfleger, K.: Irrelevant Features and the Subset Selection Problem. In: 11th International Conference on Machine Learning, pp. 121–129. Morgan Kaufmann, San Francisco (1994)Google Scholar
  3. 3.
    Baggenstoss, P.M.: Class-Specific Feature Sets in Classification. IEEE Transactions on Signal Processing, 3428–3432 (1999)Google Scholar
  4. 4.
    Baggenstoss, P.M., Niemann, H.: A Theoretically Optimal Probabilistic Classifier using Class-Specific Features. In: International Conference on Pattern Recognition (ICPR), pp. 763–768. IEEE Press, Los Alamitos (2000)Google Scholar
  5. 5.
    Baggenstoss, P.M.: Class-Specific Classifier: Avoiding the Curse of Dimensionality. IEEE Aerospace and Electronic Systems Magazine 19, 37–52 (2004)CrossRefGoogle Scholar
  6. 6.
    Baggenstoss, P.M., Beierholm, T.: Speech Music Discrimination using Class-Specific Features. In: 17th International Conference on Pattern Recognition (ICPR), pp. 379–382. IEEE Press, Los Alamitos (2004)Google Scholar
  7. 7.
    Oh, I.S., Lee, J.S., Suen, C.Y.: Analysis of Class Separation and Combination of Class-Dependent Features for Handwriting Recognition. IEEE Transactions on Pattern Analysis and Machine Intelligence 21, 1089–1094 (1999)CrossRefGoogle Scholar
  8. 8.
    Fu, X., Wang, L.: A GA-based Novel RBF Classifier with Class Dependent Features. In: Proceedings of the 2002 Congress on Evolutionary Computation CEC 2002, pp. 1890–1894. IEEE Computer Society, Los Alamitos (2002)Google Scholar
  9. 9.
    Fu, X., Wang, L.: Data mining with computational intelligence. Springer, Heidelberg (2005)zbMATHGoogle Scholar
  10. 10.
    Merz, C.J., Murphy, P.M.: UCI Repository of Machine Learning Databases. University of California at Irvine, Department of Information and Computer Science, http://ftp.ics.uci.edu/pub/machine-learning-databases/
  11. 11.
    Nanni, L.: Cluster-based Pattern Discrimination: A Novel Technique for Feature Selection. Pattern Recognition Letters 27, 682–687 (2006)CrossRefGoogle Scholar
  12. 12.
    Silva, H., Fred, A.: Feature Subspace Ensembles: A Parallel Classifier Combination Scheme using Feature Selection. In: Haindl, M., Kittler, J., Roli, F. (eds.) MCS 2007. LNCS, vol. 4472, pp. 261–270. Springer, Heidelberg (2007)CrossRefGoogle Scholar
  13. 13.
    Skurichina, M., Duin, R.P.W.: Combining Feature Subsets in Feature Selection. In: Oza, N.C., Polikar, R., Kittler, J., Roli, F. (eds.) MCS 2005. LNCS, vol. 3541, pp. 165–175. Springer, Heidelberg (2005)CrossRefGoogle Scholar
  14. 14.
    Witten, I.H., Frank, E.: Data Mining: Practical Machine Learning Tools and Techniques. Morgan Kaufmann, San Francisco (2005)zbMATHGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  • Bárbara B. Pineda-Bautista
    • 1
  • Jesús Ariel Carrasco-Ochoa
    • 1
  • José Fco. Martínez-Trinidad
    • 1
  1. 1.Computer Science Department National Institute of AstrophysicsOptics and ElectronicsPueblaMexico

Personalised recommendations