Combining feature selection methods using rank aggregation

  • Ronaldo C. Prati UFABC

Abstract


This paper investigates the use of ranking aggregation for combining feature selection methods that produces a ranked list of the features. We investigate the use of Borda, Condorcet and Kemeny methods. Results show that ranking aggregation greatly improves the quality of induced classifiers using a reduced number of features, when compared to the base feature selection methods.

References

Adali, S., Hill, B., and Magdon-Ismail, M. (2006). “The Impact of Ranker Quality on Rank Aggregation Algorithms: Information vs. Robustness”. In II Workshop on Challenges In Web Information Retrieval and Integration. IEEE Computer.

Appice, A., Ceci, M., and Flach, S. R. P. A. (2004). “Redundant Feature Elimination for Multi-Class Problems”. In Int. Conference on Machine Learning (ICML’2004).

Asuncion, A. and Newman, D. (2007). UCI machine learning repository. [link].

Demšar, J. (2006). Statistical comparisons of classifiers over multiple data sets. Journal of Machine Learning Research, 7:1–30.

Duch, W. (2006). Filter methods. In Guyon et al. (2006), chapter 3.

Dwork, C., Kumar, R., Naor, M., and Sivakumar, D. (2001). “Rank aggregation methods for the Web”. In International World Wide Web Conference, pages 613–622.

Guyon, I. and Elisseeff, A. (2003). An introduction to variable and feature selection. Journal of Machine Learning Research, 3:1157–1182.

Guyon, I., Gunn, S., Nikravesh, M., and Zadeh, L., editors (2006). Springer.

Kohavi, R. and John, G. H. (1997). Wrappers for feature subset selection. Artificial Intelligence, 97(1-2):273–324.

Kononenko, I. (1994). “Estimating Attributes: Analysis and Extensions of RELIEF”. In European Conf. on Machine Learning, volume 784 of LNCS, pages 171–182. Springer.

Lal, N., Chapelle, O., Weston, J., and Elisseeff, A. (2006). Embedded methods. In Guyon et al. (2006), chapter 5, pages 139–167.

Prati, R. C. and Monard, M. C. (2006). “Combinando métodos de seleção de subconjuntos de atributos baseados na abordagem filtro”. In I Workshop on Computational Intelligence (SBIA/SBRN/IBeramia’2006). publicado em CD-ROM.

Vafaie, H. and Jong, K. D. (1993). “Robust Feature Selection Algorithms”. In IEEE Int. Conf. on Tools with AI, pages 356–363.

Witten, I. H. and Frank, E. (2005). Data Mining: Practical Machine Learning Tools and Techniques. Morgan Kaufmann, 2 edition.
Published
2009-07-20
PRATI, Ronaldo C.. Combining feature selection methods using rank aggregation. In: NATIONAL MEETING ON ARTIFICIAL AND COMPUTATIONAL INTELLIGENCE (ENIAC), 7. , 2009, Bento Gonçalves/RS. Anais [...]. Porto Alegre: Sociedade Brasileira de Computação, 2009 . p. 21-30. ISSN 2763-9061.