JavaXCSF: the XCSF learning classifier system in Java

Authors:
Patrick O. Stalph;Martin V. Butz
Affiliations:
University of Würzburg;University of Würzburg
Venue:
ACM SIGEVOlution
Year:
2010

Citing 3
Cited 0

Classifiers that approximate functions

Natural Computing: an international journal
Classifier fitness based on accuracy

Evolutionary Computation
Function Approximation With XCS: Hyperellipsoidal Conditions, Recursive Least Squares, and Compaction

IEEE Transactions on Evolutionary Computation

Quantified Score

Hi-index	0.00

Visualization

Abstract

Learning Classifier Systems were introduced by John H. Holland and constituted one of the first genetics-based machine learning techniques. The most prominent Learning Classifier System is XCS [3]. XCS can also be used for function approximation, then called XCSF [4]. JavaXCSF is an implementation of the XCSF Learning Classifier System. It is freely available from www.coboslab.psychologie.uni-wuerzburg.de/code/ Based on previous implementations, the code was extended and heavily restructured, resulting in a flexible framework that helps newcomers to understand the basic structure, but also allows developers to realize their own ideas in the XCSF system. For a detailed documentation, please refer to [2].