IDEAS home Printed from https://ideas.repec.org/a/eee/csdana/v71y2014icp986-1000.html
   My bibliography  Save this article

(Psycho-)analysis of benchmark experiments: A formal framework for investigating the relationship between data sets and learning algorithms

Author

Listed:
  • Eugster, Manuel J.A.
  • Leisch, Friedrich
  • Strobl, Carolin

Abstract

It is common knowledge that the performance of different learning algorithms depends on certain characteristics of the data—such as dimensionality, linear separability or sample size. However, formally investigating this relationship in an objective and reproducible way is not trivial. A new formal framework for describing the relationship between data set characteristics and the performance of different learning algorithms is proposed. The framework combines the advantages of benchmark experiments with the formal description of data set characteristics by means of statistical and information-theoretic measures and with the recursive partitioning of Bradley–Terry models for comparing the algorithms’ performances. The formal aspects of each component are introduced and illustrated by means of an artificial example. Its real-world usage is demonstrated with an application example consisting of thirteen widely-used data sets and six common learning algorithms. The Appendix provides information on the implementation and the usage of the framework within the R language.

Suggested Citation

  • Eugster, Manuel J.A. & Leisch, Friedrich & Strobl, Carolin, 2014. "(Psycho-)analysis of benchmark experiments: A formal framework for investigating the relationship between data sets and learning algorithms," Computational Statistics & Data Analysis, Elsevier, vol. 71(C), pages 986-1000.
    • Handle: RePEc:eee:csdana:v:71:y:2014:i:c:p:986-1000
    DOI: 10.1016/j.csda.2013.08.007
    as

    Download full text from publisher

    File URL: http://www.sciencedirect.com/science/article/pii/S0167947313002946
    Download Restriction: Full text for ScienceDirect subscribers only.
    File URL: https://libkey.io/10.1016/j.csda.2013.08.007?utm_source=ideas
    LibKey link: if access is restricted and if your library uses this service, LibKey will redirect you to where you can use your library subscription to access this item
    ---><---

    As the access to this document is restricted, you may want to search for a different version of it.

    References listed on IDEAS

    as
    1. Douglas Critchlow & Michael Fligner, 1991. "Paired comparison, triple comparison, and ranking experiments as generalized linear models, and their implementation on GLIM," Psychometrika, Springer;The Psychometric Society, vol. 56(3), pages 517-533, September.
    2. Givens, G.H. & Beveridge, J.R. & Phillips, P.J. & Draper, B. & Lui, Y.M. & Bolme, D., 2013. "Introduction to face recognition and evaluation of algorithm performance," Computational Statistics & Data Analysis, Elsevier, vol. 67(C), pages 236-247.
    3. Achim Zeileis & Kurt Hornik, 2007. "Generalized M‐fluctuation tests for parameter instability," Statistica Neerlandica, Netherlands Society for Statistics and Operations Research, vol. 61(4), pages 488-508, November.
    4. Torti, Francesca & Perrotta, Domenico & Atkinson, Anthony C. & Riani, Marco, 2012. "Benchmark testing of algorithms for very robust regression: FS, LMS and LTS," Computational Statistics & Data Analysis, Elsevier, vol. 56(8), pages 2501-2512.
    Full references (including those not matched with items on IDEAS)

    Citations

    Citations are extracted by the CitEc Project, subscribe to its RSS feed for this item.
    as


    Cited by:

    1. Anne-Laure Boulesteix & Robert Hable & Sabine Lauer & Manuel J. A. Eugster, 2015. "A Statistical Framework for Hypothesis Testing in Real Data Comparison Studies," The American Statistician, Taylor & Francis Journals, vol. 69(3), pages 201-212, August.

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Carolin Strobl & Florian Wickelmaier & Achim Zeileis, 2011. "Accounting for Individual Differences in Bradley-Terry Models by Means of Recursive Partitioning," Journal of Educational and Behavioral Statistics, , vol. 36(2), pages 135-153, April.
    2. Grün, Bettina & Kosmidis, Ioannis & Zeileis, Achim, 2012. "Extended Beta Regression in R: Shaken, Stirred, Mixed, and Partitioned," Journal of Statistical Software, Foundation for Open Access Statistics, vol. 48(i11).
    3. Maria Teresa Alonso & Carlo Ferigato & Deimos Ibanez Segura & Domenico Perrotta & Adria Rovira-Garcia & Emmanuele Sordini, 2021. "Analysis of ‘Pre-Fit’ Datasets of gLAB by Robust Statistical Techniques," Stats, MDPI, vol. 4(2), pages 1-19, May.
    4. Jonathan Zufferey, 2016. "Investigating the migrant mortality advantage at the intersections of social stratification in Switzerland," Demographic Research, Max Planck Institute for Demographic Research, Rostock, Germany, vol. 34(32), pages 899-926.
    5. Arismendi, Juan C. & Broda, Simon, 2017. "Multivariate elliptical truncated moments," Journal of Multivariate Analysis, Elsevier, vol. 157(C), pages 29-44.
    6. Lisa Dandolo & Klaus Telkmann & Christina Hartig & Sophie Horstmann & Sara Pedron & Lars Schwettmann & Peter Selsam & Alexandra Schneider & Gabriele Bolte & on behalf of the INGER Study Group, 2023. "Do Multiple Sex/Gender Dimensions Play a Role in the Association of Green Space and Self-Rated Health? Model-Based Recursive Partitioning Results from the KORA INGER Study," IJERPH, MDPI, vol. 20(7), pages 1-23, March.
    7. Selin Ahipaşaoğlu, 2015. "Fast algorithms for the minimum volume estimator," Journal of Global Optimization, Springer, vol. 62(2), pages 351-370, June.
    8. Payton J. Jones & Patrick Mair & Thorsten Simon & Achim Zeileis, 2020. "Network Trees: A Method for Recursively Partitioning Covariance Structures," Psychometrika, Springer;The Psychometric Society, vol. 85(4), pages 926-945, December.
    9. Brunori, Paolo & Davillas, Apostolos & Jones, Andrew M. & Scarchilli, Giovanna, 2022. "Model-based Recursive Partitioning to Estimate Unfair Health Inequalities in the United Kingdom Household Longitudinal Study," Journal of Economic Behavior & Organization, Elsevier, vol. 204(C), pages 543-565.
    10. Edgar Merkle & Achim Zeileis, 2013. "Tests of Measurement Invariance Without Subgroups: A Generalization of Classical Methods," Psychometrika, Springer;The Psychometric Society, vol. 78(1), pages 59-82, January.
    11. Jan J. J. Groen & George Kapetanios & Simon Price, 2013. "Multivariate Methods For Monitoring Structural Change," Journal of Applied Econometrics, John Wiley & Sons, Ltd., vol. 28(2), pages 250-274, March.
    12. Seibold Heidi & Zeileis Achim & Hothorn Torsten, 2016. "Model-Based Recursive Partitioning for Subgroup Analyses," The International Journal of Biostatistics, De Gruyter, vol. 12(1), pages 45-63, May.
    13. Hatzinger, Reinhold & Dittrich, Regina, 2012. "prefmod: An R Package for Modeling Preferences Based on Paired Comparisons, Rankings, or Ratings," Journal of Statistical Software, Foundation for Open Access Statistics, vol. 48(i10).
    14. Susan Athey & Julie Tibshirani & Stefan Wager, 2016. "Generalized Random Forests," Papers 1610.01271, arXiv.org, revised Apr 2018.
    15. Edgar Merkle & Jinyan Fan & Achim Zeileis, 2014. "Testing for Measurement Invariance with Respect to an Ordinal Variable," Psychometrika, Springer;The Psychometric Society, vol. 79(4), pages 569-584, October.
    16. Charles G. Renfro, 2009. "The Practice of Econometric Theory," Advanced Studies in Theoretical and Applied Econometrics, Springer, number 978-3-540-75571-5, February.
    17. Baishuai Zuo & Chuancun Yin & Jing Yao, 2023. "Multivariate range Value-at-Risk and covariance risk measures for elliptical and log-elliptical distributions," Papers 2305.09097, arXiv.org.
    18. Yu-Shan Shih & Kuang-Hsun Liu, 2019. "Regression trees for detecting preference patterns from rank data," Advances in Data Analysis and Classification, Springer;German Classification Society - Gesellschaft für Klassifikation (GfKl);Japanese Classification Society (JCS);Classification and Data Analysis Group of the Italian Statistical Society (CLADAG);International Federation of Classification Societies (IFCS), vol. 13(3), pages 683-702, September.
    19. Sies Aniek & Van Mechelen Iven, 2017. "Comparing Four Methods for Estimating Tree-Based Treatment Regimes," The International Journal of Biostatistics, De Gruyter, vol. 13(1), pages 1-20, May.
    20. Baishuai Zuo & Chuancun Yin, 2022. "Multivariate doubly truncated moments for generalized skew-elliptical distributions with application to multivariate tail conditional risk measures," Papers 2203.00839, arXiv.org.

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:eee:csdana:v:71:y:2014:i:c:p:986-1000. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: Catherine Liu (email available below). General contact details of provider: http://www.elsevier.com/locate/csda .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.