An Alternative to ROC and AUC Analysis of Classifiers

  • Frank Klawonn
  • Frank Höppner
  • Sigrun May
Part of the Lecture Notes in Computer Science book series (LNCS, volume 7014)


Performance evaluation of classifiers is a crucial step for selecting the best classifier or the best set of parameters for a classifier. The misclassification rate of a classifier is often too simple because it does not take into account that misclassification for different classes might have more or less serious consequences. On the other hand, it is often difficult to specify exactly the consequences or costs of misclassifications. ROC and AUC analysis try to overcome these problems, but have their own disadvantages and even inconsistencies. We propose a visualisation technique for classifier performance evaluation and comparison that avoids the problems of ROC and AUC analysis.


Receiver Operating Characteristic Receiver Operating Characteristic Curve Pareto Front Optimal Threshold Area Under Curve 
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.
    Hand, D.: Measuring classifier performance: a coherent alternative to the area under the ROC curve. Machine Learning 77, 103–123 (2009)CrossRefGoogle Scholar
  2. 2.
    Kohavi, R.: A study of cross-validation and bootstrap for accuracy estimation and model selection. In: Proceedings of the Fourteenth International Joint Conference on Artificial Intelligence, pp. 1137–1143. Morgan Kaufmann, San Mateo (1995)Google Scholar
  3. 3.
    Obuchowsky, N., Lieber, M., Wians Jr., F.: ROC curves in clinical chemistry: Uses, misuses and possible solutions. Clinical Chemistry 50, 1118–1125 (2004)CrossRefGoogle Scholar
  4. 4.
    Søreide, K.: Receiver-operating characteristic (ROC) curve analysis in diagnostic, prognostic and predictive biomarker research. Clinical Pathology 62, 1–5 (2009)CrossRefGoogle Scholar
  5. 5.
    Berthold, M., Borgelt, C., Höppner, F., Klawonn, F.: Guide to Intelligent Data Analysis: How to Intelligently Make Sense of Real Data. Springer, London (2010)CrossRefzbMATHGoogle Scholar
  6. 6.
    Hand, D., Mannila, H., Smyth, P.: Principles of Data Mining. MIT Press, Cambridge (2001)Google Scholar
  7. 7.
    Provost, F., Fawcett, T., Kohavi, R.: The case against accuracy estimation for comparing induction algorithms. In: Proceedings of the 15th International Conference on Machine Learning (1998)Google Scholar
  8. 8.
    Mossman, D.: Three-way ROCs. Medical Decision Making 19, 78–89 (1999)CrossRefGoogle Scholar
  9. 9.
    Hand, D., Till, R.: A simple generalisation of the area under the ROC curve for multiple class classification problems. Machine Learning 45, 171–186 (2001)CrossRefzbMATHGoogle Scholar
  10. 10.
    Li, J., Fine, J.: ROC analysis with multiple classes and multiple tests: methodology and its application in microarray studies. Biostatistics 9, 566–576 (2008)CrossRefzbMATHGoogle Scholar
  11. 11.
    Adams, N., Hand, D.: Comparing classifiers when the misallocation costs are uncertain. Pattern Recognition 32, 1139–1147 (1999)CrossRefGoogle Scholar
  12. 12.
    Drummond, C., Holte, R.: Explicitly representing expected cost: An alternative to ROC representation. In: Proc. Sixth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pp. 198–207. ACM Press, New York (2000)CrossRefGoogle Scholar
  13. 13.
    Drummond, C., Holte, R.: Cost curves: An improved method for visualizing classifier performance. Machine Learning 65, 95–130 (2006)CrossRefGoogle Scholar
  14. 14.
    Hernández-Orallo, J., Flach, P., Ferri, C.: Brier curves: a new cost-based visualisation of classifier performance. In: Getoor, L., Scheffer, T. (eds.) Proc. 28th International Conference on Machine Learning (ICML 2011), pp. 585–592. ACM, New York (2011)Google Scholar
  15. 15.
    Turney, P.: Cost-sensitive classification: Empirical evaluation of a hybrid genetic decision tree induction algorithm. Journal of Artificial Intelligence Research 2, 369–409 (1995)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2011

Authors and Affiliations

  • Frank Klawonn
    • 1
    • 2
  • Frank Höppner
    • 1
  • Sigrun May
    • 3
  1. 1.Department of Computer ScienceOstfalia University of Applied SciencesWolfenbuettelGermany
  2. 2.Bioinformatics and StatisticsHelmholtz Centre for Infection ResearchBraunschweigGermany
  3. 3.Biological Systems AnalysisHelmholtz Centre for Infection ResearchBraunschweigGermany

Personalised recommendations