A Comparison of One-Class Classifiers for Novelty Detection in Forensic Case Data

  • Frédéric Ratle
  • Mikhail Kanevski
  • Anne-Laure Terrettaz-Zufferey
  • Pierre Esseiva
  • Olivier Ribaux
Part of the Lecture Notes in Computer Science book series (LNCS, volume 4881)


This paper investigates the application of novelty detection techniques to the problem of drug profiling in forensic science. Numerous one-class classifiers are tried out, from the simple k-means to the more elaborate Support Vector Data Description algorithm. The target application is the classification of illicit drugs samples as part of an existing trafficking network or as a new cluster. A unique chemical database of heroin and cocaine seizures is available and allows assessing the methods. Evaluation is done using the area under the ROC curve of the classifiers. Gaussian mixture models and the SVDD method are trained both with and without outlier examples, and it is found that providing outliers during training improves in some cases the classification performance. Finally, combination schemes of classifiers are also tried out. Results highlight methods that may guide the profiling methodology used in forensic analysis.


Receiver Operating Characteristic Curve Gaussian Mixture Model Target Data Novelty Detection Support Vector Data Description 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Esseiva, P., Dujourdy, L., Anglada, F., Taroni, F., Margot, P.: A methodology for illicit heroin seizures comparison in a drug intelligence perspective using large databases. Forensic Science International 132, 139–152 (2003)CrossRefGoogle Scholar
  2. 2.
    Esseiva, P., Anglada, F., Dujourdy, L., Taroni, F., Margot, P., Du Pasquier, E., Dawson, M., Roux, C., Doble, P.: Chemical profiling and classification of illicit heroin by principal component analysis. Calculation of inter sample correlation and artificial neural networks, Talanta 67, 360–367 (2005)Google Scholar
  3. 3.
    Ratle, F., Terrettaz-Zufferey, A.L., Kanevski, M., Esseiva, P., Ribaux, O.: Pattern analysis in illicit heroin seizures: a novel application of machine learning algorithms. In: Proc. of the 14th European Symposium on Artificial Neural Networks, d-side publi. (2006)Google Scholar
  4. 4.
    Ratle, F., Terrettaz-Zufferey, A.L., Kanevski, M., Esseiva, P., Ribaux, O.: Learning manifolds in forensic data. In: Proc. of the 16th Int. Conf. on Artificial Neural Networks, Springer, Heidelberg (2006)Google Scholar
  5. 5.
    Tong, C., Svetnik, V.: Novelty detection in mass spectral data using a support vector machine method. In: Proc. of Interface 2002 (2002)Google Scholar
  6. 6.
    Markou, M., Singh, S.: Novelty detection: a review - part 1: statistical approaches. Signal Processing 83, 2481–2497 (2003)CrossRefGoogle Scholar
  7. 7.
    Markou, M., Singh, S.: Novelty detection: a review - part 2: neural network based approaches. Signal Processing 83, 2499–2521 (2003)CrossRefGoogle Scholar
  8. 8.
    Marsland, S.: Novelty detection in learning systems. Neural Computing Surveys 3, 157–195 (2003)Google Scholar
  9. 9.
    Tax, D.M.J.: One-class classification, Ph.D. thesis, University of Amsterdam (2001)Google Scholar
  10. 10.
    Tax, D.M.J., Duin, R.P.W.: Support Vector Data Description. Machine Learning 54, 45–66 (2004)zbMATHCrossRefGoogle Scholar
  11. 11.
    Schölkopf, B., Williamson, R., Smola, A., Shawe-Taylor, J., Platt, J.: Support vector method for novelty detection. Advances in Neural Information Processing Systems 12 (2000)Google Scholar
  12. 12.
    Tax, D.M.J.: Combining one-class classifiers. In: Kittler, J., Roli, F. (eds.) MCS 2001. LNCS, vol. 2096, Springer, Heidelberg (2001)Google Scholar
  13. 13.
    Guéniat, O., Esseiva, P.: Le Profilage de l’Héroïne et de la Cocaïne, Presses polytechniques et universitaires romandes, Lausanne (2005)Google Scholar
  14. 14.
    Tax, D.M.J.: DDtools, the Data Description Toolbox for Matlab, version 1.5.7 (2007)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2007

Authors and Affiliations

  • Frédéric Ratle
    • 1
  • Mikhail Kanevski
    • 1
  • Anne-Laure Terrettaz-Zufferey
    • 2
  • Pierre Esseiva
    • 2
  • Olivier Ribaux
    • 2
  1. 1.Institute of Geomatics and Risk Analysis, Faculty of Earth and Environmental Sciences, University of Lausanne, CH-1015Switzerland
  2. 2.School of Criminal Sciences, Faculty of Law, University of Lausanne, CH-1015Switzerland

Personalised recommendations