Advertisement

Variable Selection in Industrial Datasets Using Pareto Genetic Programming

  • Guido Smits
  • Arthur Kordon
  • Katherine Vladislavleva
  • Elsa Jordaan
  • Mark Kotanchek
Part of the Genetic Programming book series (GPEM, volume 9)

Abstract

This chapter gives an overview, based on the experience from the Dow Chemical Company, of the importance of variable selection to build robust models from industrial datasets. A quick review of variable selection schemes based on linear techniques is given. A relatively simple fitness inheritance scheme is proposed to do nonlinear sensitivity analysis that is especially effective when combined with Pareto GP. The method is applied to two industrial datasets with good results.

Key words

Genetic programming symbolic regression variable selection pareto GP 

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. Banzhaf, W., Nordin, P., Keller, R., and Francone, F. (1998). Genetic Programming: An Introduction, San Francisco, CA: Morgan Kaufmann.Google Scholar
  2. Cherkassky V, Mulier, F., 1998, “Learning from data, Concepts, Theory and Methods”, Wiley Interscience, ISBN 0-471-15493-8.Google Scholar
  3. Draper, N. R. and Smith, H. (1981) Applied Regression Analysis, Second Edition, New York, NY: Wiley.Google Scholar
  4. Eriksson, L., Johansson, E., Wold, N., and Wold, S. (2001). Multi and Megavariate Data Analysis: Principles and Applications, Umea, Sweden, Umetrics Academy.Google Scholar
  5. Francone, F. et al (2004). Discrimination of Unexploded Ordnance from Clutter Using Linear Genetic Programming, Genetic and Evolutionary Computation Conference, Late Breaking Papers.Google Scholar
  6. Gilbert, R.J., Goodacre, R., Shann, B., Taylor, J., Rowland, J.J. and Kell, D.B., Genetic Programming-Based Variable Selection for High-Dimensional Data, in J.R. Koza et al., editors, Genetic Programming 1998: Proceedings of the Third Annual Conference (GP-98), Madison, WI 22–25 July 1998, Morgan Kaufmann, San Fransisco, CA.Google Scholar
  7. Ohnson, H.E, Gilbert, R.J., Winson, M.K., Goodacre, R., Smith, A.R., Rowand, J.J., Hall, M.A. and Kell, D.B. Explanatory Analysis of he Metabolome Using Genetic Programming of Simple, Interpretable Rules, in Genetic Programming and Evolvable Machines, Vol 1 (2000)Google Scholar
  8. Jordaan, E., Kordon, A., Smits, G., and Chiang L. (2004). Robust Inferential Sensors based on Ensemble of predictors generated by Genetic Programming, In Proceedings of PPSN 2004, pp. 522–531, Birmingham, UK.Google Scholar
  9. Kordon, A., Smits, G., Kalos, A., and Jordaan, E. (2003). Robust Soft Sensor Development Using Genetic Programming, In Nature-Inspired Methods in Chemometrics, (R. Leardi-Editor), Amsterdam: ElsevierGoogle Scholar
  10. Kotanchek, M, Smits, G. and Kordon, A. (2003). Industrial Strength Genetic Programming, In Genetic Programming Theory and Practice, pp 239–258, R. Riolo and B. Worzel (Eds), Boston, MA: Kluwer.Google Scholar
  11. Koza, J. (1992). Genetic Programming: On the Programming of Computers by Means of Natural Selection, Cambridge, MA: MIT Press.Google Scholar
  12. RML Technologies, Inc. (2002) Discipulus Owner’s Manual.Google Scholar
  13. Saltelli A., Chan K., and Scott E. (2001). Sensitivity Analysis, Baffins Lane, Chichester, UK: Wiley.Google Scholar
  14. Smits, G. and Kotanchek. (2004), Pareto-Front Exploitation in Symbolic Regression, Genetic Programming Theory and Practice, pp 283–300, R. Riolo and B. Worzel (Eds), Boston, MA: Kluwer.Google Scholar

Copyright information

© Springer Science+Business Media, Inc. 2006

Authors and Affiliations

  • Guido Smits
    • 1
  • Arthur Kordon
    • 2
  • Katherine Vladislavleva
    • 1
  • Elsa Jordaan
    • 1
  • Mark Kotanchek
    • 3
  1. 1.Dow BeneluxTerneuzenThe Netherlands
  2. 2.The Dow Chemical CompanyFreeport
  3. 3.The Dow Chemical CompanyMidland

Personalised recommendations