This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Feature Selection for Multiclass Discrimination via Mixed-Integer Linear Programming
June 2003 (vol. 25 no. 6)
pp. 779-783

Abstract—We reformulate branch-and-bound feature selection employing L_\infty or particular L_p metrics, as mixed-integer linear programming (MILP) problems, affording convenience of widely available MILP solvers. These formulations offer direct influence over individual pairwise interclass margins, which is useful for feature selection in multiclass settings.

[1] R.O. Duda, P.E. Hart, and D.G. Stork, Pattern Classification. second ed. New York: John Wiley and Sons, Inc., 2001.
[2] C.M. Bishop, Neural Networks for Pattern Recognition. New York: Oxford Univ. Press, 1995.
[3] M. Dash and H. Liu, Feature Selection for Classification Intelligent Data Analysis, vol. 1, pp. 131-156, 1997.
[4] M. Kudo and J. Sklansky, Comparison of Algorithms That Select Features for Pattern Classifiers Pattern Recognition, vol. 33, pp. 25-41, 2000.
[5] A. Jain and D. Zongker, Feature Selection: Evaluation, Application, and Small Sample Performance IEEE Trans. Pattern Analysis and Machine Intelligence, vol. 19, no. 2, pp. 153-158, Feb. 1997.
[6] J. Glen, Classification Accuracy in Discriminant Analysis: A Mixed Integer Programming Approach J. Operational Research Soc., vol. 52, pp. 328-339, 2001.
[7] D.J. Hand, Discrimination and Classification, chapter 6. John Wiley and Sons, 1981.
[8] T.M. Cover and J.M.V. Campenhout, On the Possible Orderings in the Measurement Selection Problem IEEE Trans. Systems, Man, and Cybernetics, vol. 7, no. 9, pp. 657-661, 1977.
[9] P. Narendra and K. Fukunaga, A Branch and Bound Algorithm for Feature Subset Selection IEEE Trans. Computers, vol. 26, no. 9, pp. 917-922, Sept. 1977.
[10] T.M. Mitchell, Machine Learning. New York: McGraw-Hill, 1997.
[11] J.T. Tou and R.C. Gonzalez, Pattern Recognition Principles. London: Addison- Wesley, 1974.
[12] L. Bruzzone and S.B. Serpico, A Technique for Feature Selection in Multiclass Problems Int'l J. Remote Sensing, vol. 21, no. 3, pp. 549-563, 2000.
[13] L.A. Wolsey, Integer Programming. New York: John Wiley and Sons, Inc., 1998.
[14] R.J. Vanderbei, Linear Programming, second ed. Boston: Kluwer Academic Publishers, 2001.
[15] J. Mitchell and E. Lee, Branch-and-Bound Methods for Integer Programming Encyclopedia of Optimization, Kluwer Academic Publishers, 2001.
[16] J.T. Linderoth and M.W.P. Savelsbergh, A Computational Study of Branch and Bound Search Strategies for Mixed Integer Programming INFORMS J. Computing, vol. 11, pp. 173-187, 1999.
[17] C. Blake and C. Merz, UCI Repository of Machine Learning Databases http://www.ics.uci.edu/~mlearnMLRepository.html , 1998.
[18] R. Duin, PRTools 3, A Matlab Toolbox for Pattern Recognition. Delft Univ. of Tech nology, 2000.

Index Terms:
Feature selection, discrimination, classification, mixed-integer linear programming, branch-and-bound.
Citation:
Frank J. Iannarilli Jr., Paul A. Rubin, "Feature Selection for Multiclass Discrimination via Mixed-Integer Linear Programming," IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 25, no. 6, pp. 779-783, June 2003, doi:10.1109/TPAMI.2003.1201827
Usage of this product signifies your acceptance of the Terms of Use.