• DocumentCode
    595083
  • Title

    Null QQ plots: A simple graphical alternative to significance testing for the comparison of classifiers

  • Author

    Berrar, D.

  • Author_Institution
    Tokyo Inst. of Technol., Yokohama, Japan
  • fYear
    2012
  • fDate
    11-15 Nov. 2012
  • Firstpage
    1852
  • Lastpage
    1855
  • Abstract
    The evaluation of machine learning algorithms is commonly based on statistical significance tests. However, the suitability of such tests is often questionable. We propose null QQ plots as a simple yet powerful graphical alternative to significance testing. Using ten benchmark data sets, we demonstrate that these plots concisely summarize the essential results from a comparative classification study, while they are easy to produce and interpret.
  • Keywords
    learning (artificial intelligence); pattern classification; statistical testing; comparative classification study; graphical alternative; machine learning algorithms; null QQ plots; significance testing; statistical significance tests; ten benchmark data sets; Accuracy; Data models; Machine learning; Machine learning algorithms; Single photon emission computed tomography; Sonar; Testing;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Pattern Recognition (ICPR), 2012 21st International Conference on
  • Conference_Location
    Tsukuba
  • ISSN
    1051-4651
  • Print_ISBN
    978-1-4673-2216-4
  • Type

    conf

  • Filename
    6460514