A Generic Approach to an Object-Oriented Learning Classifier System Library
Matthias Meyer, Klaus Hufschlag · RePEc: Research Papers in Economics · 2006
Learning Classifier Systems (LCS) have gained popularity in the realm of social science simulation. However, when it comes to actually constructing a LCS for a particular modelling purpose, it seems that every researcher must currently "reinvent the wheel". Taking this situation as a starting point, the objective of this paper is to present the basic ideas behind a LCS library, which can relieve simulation researchers of some of the technical work and can provide a generic structure for modelling LCS. The library is based on a strictly object-oriented approach. This provides flexibility in the process of constructing a LCS for a specific modelling purpose and encourages experimentation with various different assumptions. The paper is supported with examples based on experience in using the library.