Post-processing clustering to reduce XCS variability

XCS is a stochastic algorithm, so it does not guarantee to produce the same results when run with the same input. When interpretability matters, obtaining a single, stable result is important. We propose an algorithm to join the rules produced from many XCS runs, based on a measure of distance between rules. We also suggest a general definition for such a measure, and show the results obtained on a complex data set. Categories and Subject Descriptors I.2.6 [Artificial Intelligence]: Learning—Learning classifier systems General Terms Algorithms Keywords Learning classifier systems, clustering

• Type: Conference
• Year: 2005
• Where: GECCO
• Authors: Flavio Baronti, Alessandro Passaro, Antonina Starita
• Keywords: Classifier Systems, GECCO 2005, Stochastic Algorithm, Terms Algorithms Keywords