Mathematical Geology

, Volume 39, Issue 3, pp 307–319 | Cite as

Support Vector Machines for Classification of Aggregates by Means of IR-Spectra

  • Vera Hofer
  • Juergen Pilz
  • Thorgeir S. Helgason


The increasing physical and technical demands placed on construction materials, especially as they are being used more and more up to the limits of their mechanical strength, has led the aggregates industry to search for more efficient methods of quality control. Information from theoretical work on rock spectra in near-infrared and mid-infrared light as well as achievements gained in signal processing can all be used to improve quality control in an economically acceptable manner. As engineering properties of aggregates are to a great extent determined by the petrological composition of the rock aggregates, the question is, whether a statistical classification rule for identification of rock aggregates can be developed. However, the classification of rocks is complicated by the fact that the optical behavior of minerals forming the rock often appears muted. In addition, minor constituents may dominate the spectrum. Furthermore, the relevant spectra form high dimensional data that are extremely difficult to analyze statistically, especially when curves are very similar. In this paper, support vector machines for classification of rock spectra are investigated, since they are appropriate in classifying highly dimensional data such as spectra.


Wavelets Principal component analysis Partial least squares Directed acyclic graph 



Directed acyclic graph


Support vector machines for multi-groups classification using DAG


Mid infrared light


Near infrared light


Data without any transformation


Data after PCA


Data after PLS


Principal component


Principal component analysis


Partial least squares


Support vector machines


Wavelet transformed data with subsequent PCA


Wavelet transformed data with subsequent PLS


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Abrahamsson, C., Johansson, J., Sparén, A., and Lindgren, F., 2003, Comparison of different variable selection methods conducted on NIR transmission measurements on intact tablets: Chemom. Intell. Lab. Syst., v. 69, p. 3–12. CrossRefGoogle Scholar
  2. Burges, C.J.C., 1998, A tutorial on support vector machines for pattern recognition: Knowl. Discov. Data Min., v. 2, no. 2, p. 121–167. CrossRefGoogle Scholar
  3. Cook, R.D., and Lee, H., 1999, Dimension reduction in binary response regression: Am. Stat. Assoc., v. 94, p. 1187–1200. CrossRefGoogle Scholar
  4. Cristianini, N., and Shawe-Taylor, J., 2004, An Introduction to Support Vector Machines: Cambridge University Press, Cambridge, 189 p. Google Scholar
  5. EUREKA, 2004, PETROSCOPE—An optical analyser for construction aggregates and rocks: Project no. 2569, Announced 28 June 2001, EUREKA, Brussels. Google Scholar
  6. EUREKA, 2005, PETROSCOPE II: Project no. 3665, Announced 1 June 2005, EUREKA, Brussels. Google Scholar
  7. Hastie, T., Tibshirani, R., and Buja, A., 1995, Penalized discriminant analysis: Ann. Stat., v. 23, p. 73–102. Google Scholar
  8. Hastie, T.J, Tibshirani, R.J, and Friedman, J., 2001, The Elements of Statistical Learning, Data Mining, Inference and Prediction: Springer, New York, 533 p. Google Scholar
  9. Hastie, T.J., Rosset, S., Tibshirani, R., and Zhu, J., 2004, The eentire regularization path for the support vector machine: J. Mach. Learn. Res., v. 5, p. 1391–1415. Google Scholar
  10. Hofer, V., Pilz, J., and Helgason, T.S., 2006, Statistical classification of aggregates by means of NIR and MIR spectra: Math. Geol., v. 38, no. 7, online available, DOI:  10.1007/s11004-006-9052-x.
  11. Hsu, C.-W., and Lin, C.-J., 2002, A comparison of methods for multi-class support vector machines: IEEE Trans. Neural Netw., v. 13, no. 2, p. 415–425. CrossRefGoogle Scholar
  12. Hsu, C.-W., Chang, C.-C., and Lin, C.-J., 2003, A practical guide to support vector classification,Technical Report, Department of Computer Science and Information Engineering, National Taiwan University (
  13. Hunt, G.R., 1982, Spectroscopic properties of rocks and minerals, In: Carmichael, R.S., ed., Handbook of Physical Properties of Rocks: CRC Press, Boca Raton, p. 295–385. Google Scholar
  14. Lee, Y., Lin, Y, Wahba, G., 2004, Multicategory support vector machines:theory and application to the classification of microarray data and satellite radiance data: Am. Stat. Assoc., v. 99, no. 465, p. 67–81. CrossRefGoogle Scholar
  15. Li, K.-C., Aragon, Y., Shedden, K., and Agnan, C.T., 2003, Dimension reduction for multivariate responses: Am. Stat. Assoc., v. 98(461), p. 99–109. CrossRefGoogle Scholar
  16. Jolliffe, I.T., 1986, Principal Component Analysis: Springer Series in Statistics: Springer, New York, 487 p. Google Scholar
  17. McLachlan, G.J., 1992, Discriminant Analysis and Statistical Pattern Recognition, Wiley, New York, 526 p. Google Scholar
  18. Platt, J., Cristianini, N., and Shawe-Taylor, J., 2000, Large margin DAGs for multiclass classification: Adv. Neural Inf. Proc. Syst., v. 12, p. 547–553. Google Scholar
  19. Ramsay, J.O., and Silverman, B.W., 2005, Functional Data Analysis: Springer Series in Statistics: Springer, New York, 310 p. Google Scholar
  20. Schmidt, W., 1994, Optische Spektroskopie, Eine Einführung für Naturwissenschaftler und Techniker, Wiley–VCH, New York, 426 p. Google Scholar
  21. Smith, M.R., and Fookes, P.G., 2001, Aggregates—Sand, Gravel and Crushed Rock Aggregates for Construction Purposes, 3 rd ed.: Geol. Soc. London, London, 339 p. Google Scholar
  22. Teppola, P., and Minkkinen, 2000, Wavelet-PLS regression models for both exploratory data analysis and process monitoring: J. Chemom., v. 14, p. 383–399. CrossRefGoogle Scholar
  23. Vannucci, M., Brown, P.J., and Fearn, T., 2003, A decision theoretical approach to wavelet regression on curves with a high number of regressors: J. Stat. Plan. Inference, v. 112, p. 195–212. CrossRefGoogle Scholar
  24. Vidakovic, B., 1999, Statistical Modeling by Wavelets: Wiley, New York, 382 p. Google Scholar
  25. Vidakovic, B., and Müller, P., 1999, An introduction to wavelets, in Müller, P. and Vidakovic, B., eds., Bayesian Inference in Wavelet-Based Models, Springer, New York, 394 p. Google Scholar
  26. VTT/Electronics, 2001, Petroscope prestudy: Confidential Report to Petromodel Ltd, VTT, Oulu, Finland. Google Scholar

Copyright information

© International Association for Mathematical Geology 2007

Authors and Affiliations

  • Vera Hofer
    • 1
  • Juergen Pilz
    • 2
  • Thorgeir S. Helgason
    • 3
  1. 1.Department of Statistics and Operations ResearchKarl-Franzens UniversityGrazAustria
  2. 2.Department of MathematicsUniversity of KlagenfurtKlagenfurtAustria
  3. 3.Petromodel ehfReykjavikIceland

Personalised recommendations