Talk:Accuracy paradox

From Wikipedia, the free encyclopedia
Jump to: navigation, search

Potential sources[edit]

  • Woehr, David J., "Understanding frame-of-reference training: The impact of training on the recall of performance information.", Journal of Applied Psychology. Vol 79(4), Aug 1994, 525-534. doi:10.1037/0021-9010.79.4.525
  • Fernandes, J.A. / Irigoien, X. / Goikoetxea, N. / Lozano, J.A. / Inza, I. / Perez, A. / Bode, A., "Fish recruitment prediction, using robust supervised classification methods", Ecological Modelling, 221 (2), p.338-352, Jan 2010, doi:10.1016/j.ecolmodel.2009.09.020

Similarly-named paradoxes[edit]

Solutions to the accuracy paradox[edit]

  • The correct solution to the 'paradox' is to think about the information gained by using the model instead of the accuracy. See Information_gain_in_decision_trees for an example of how to measure the information (conditional entropy). This concept worked out for the example currently on the page would add value to this article — Preceding unsigned comment added by Bscan (talkcontribs) 19:28, 29 October 2013 (UTC)