Method of Wavelength Selection for Partial Least Squares

(Note: The full text of this document is currently only available in the PDF Version )

Scott D. Osborne, Rainer Künnemeyer, Scott D. Osborne and Robert B. Jordan


Abstract

A new method for the selection of wavelengths from near infrared spectra using partial least squares(PLS) analysis is presented. The method aims to find wavelengths that produce significant improvements in PLS prediction accuracy over using all wavelengths. The method is based on data splitting and evaluation of the appropriate prediction errors. Analysis of interactance spectra of kiwifruit using three evaluation criteria are compared with the results obtained from full spectrum analysis and with the recently proposed feature selection method. Using the recommended criterion, the method was found to produce models with lower standard errors than the optimum model obtained using the feature selection method 87% of the time. Properties of initiating the search method from starting points selected by three procedures are compared and recommendations are given for selecting the initial wavelengths. The new search method also has a low probability of obtaining significant correlations through chance.


References

  1. H. Wold, Multivariate Analysis, Academic Press, New York, 1966, p. 391 Search PubMed.
  2. P. Geladi and B. R. Kowalski, Anal. Chim. Acta, 1986, 185, 1 CrossRef CAS.
  3. T. Naes, C. Irgens and H. Martens, Appl. Stat., 1986, 35, 195.
  4. W. Lindberg, J. A. Persson and S. Wold, Anal. Chem., 1983, 55, 643 CrossRef CAS.
  5. P. J. Young, SIAM J. Sci. Comput., 1994, 15, 225 Search PubMed.
  6. N. R. Draper and H. Smith, Applied Regression Analysis, Wiley, New York, 2nd edn., 1981 Search PubMed.
  7. F. Lindgren, P. Geladi, S. Rännar and S. Wold, J. Chemom., 1994, 8, 349.
  8. A. G. Frenich, D. Jouan-Rimbaud, D. L. Massart, S. Kuttatharmmakul, M. Martínez Galera and J. L. Martínez Vidal, Analyst, 1995, 120, 2787 RSC.
  9. I. E. Frank, Chemom. Intell. Lab. Syst., 1987, 1, 233 CrossRef.
  10. M. Stone, J. Royal Stat. Soc., 1974, Series B, 111 Search PubMed.
  11. R. Leardi, R. Boggia and M. Terrile, J. Chemom, 1992, 6, 267 CAS.
  12. C. B. Lucasius, M. L. M. Beckers and G. Kateman, Anal. Chim. Acta, 1994, 286, 135 CrossRef CAS.
  13. R. Leardi, J. Chemom., 1994, 8, 65 CAS.
  14. J. H. Kalivas, N. Roberts and M. Sutter, Anal. Chem., 1989, 61, 2024 CrossRef CAS.
  15. R. P. Picard and K. N. Berk, Am. Stat., 1990, 44, 140.
  16. P. J. Brown, J. Chemom., 1994, 6, 151.
  17. S. D. Osborne, R. B. Jordan and R. Künnemeyer, Using Near Infrared (NIR) Light to Estimate the Soluble Solids and Dry Matter Content of Kiwifruit, Presented at the PH'96 International Postharvest Science Conference, Taupo, New Zealand, August 1996 Search PubMed.
  18. R. B. Jordan, S. D. Osborne, R. Künnemeyer and R. J. Seelye, Harvest Time Prediction of Eating Time Properties of Kiwifruit Using NIR Transmission, Proceedings from the Sensors for Nondestructive Testing International Conference, Orlando, Florida, February 1997, p. 101 Search PubMed.
  19. B. M. Wise, PLS_Toolbox, analysis routines written for Matlab software available over the Internet, Center for Process Analytical Chemistry and Department of Chemical Engineering BF-10, University of Washington, Seattle, Washington 98195, bmw@chevax.chema.washington.edu Search PubMed.
  20. M. Clark and R. D. Cramer III, Quant. Struct.-Act. Relat., 1993, 12, 137 Search PubMed.
  21. D. Jouan-Rimbaud, B. Walczak, D. L. Massart, I. R. Last and K. A. Prebble, Anal. Chim. Acta, 1995, 304, 285 CrossRef CAS.
Click here to see how this site uses Cookies. View our privacy policy here.