The XCSF classifier system in Java

Abstract

After many inquiries by various researchers in the field, I finally got around to completely restructure my XCS Java code released in 2000 (!) and was able to make a new version available. Due the apparent lack of publicly available real-valued XCS implementations and the struggle of many to implement it from scratch, I invested the time and put out a real-valued XCS implementation for function approximation. The code is available from Martin Pelikan's MEDAL lab webpage (http://medal.cs.umsl.edu/files/XCSFJava1.1.zip). It includes the Java API, which should give a general idea of how the code works and how classes communicate with each other. Moreover, there is a short MEDAL report available that gives further code details and also points out what needs to be done to adjust or enhance the code for your individual research needs (http://medal.cs.umsl.edu/files/2007008.pdf). This note is meant to spread the good news and to give a short idea of what the code is capable of.

References

1. M. V. Butz, P. L. Lanzi, S. W. Wilson. Function Approximation with XCS: Hyperellipsoidal Conditions, Recursive Least Squares, and Compaction IEEE Transactions on Evolutionary Computation. (in press).

Index Terms

• Computing methodologies

  • Machine learning

    • Learning paradigms

      • Supervised learning

        • Supervised learning by classification

    • Machine learning approaches

      • Classification and regression trees

• Software and its engineering

  • Software notations and tools

    • General programming languages

      • Language types