Compact Rulesets from XCSI

  • Authors:
  • Stewart W. Wilson

  • Affiliations:
  • -

  • Venue:
  • IWLCS '01 Revised Papers from the 4th International Workshop on Advances in Learning Classifier Systems
  • Year:
  • 2001

Quantified Score

Hi-index 0.00

Visualization

Abstract

An algorithm is presented for reducing the size of evolved classifier populations. On the Wisconsin Breast Cancer dataset, the algorithm produced compact rulesets substantially smaller than the populations, yet performance in cross-validation tests was nearly unchanged. Classifiers of the rulesets expressed readily interpretable knowledge about the dataset that should be useful to practitioners.