Abstract: |
XCS is widely accepted as one of the most reliable Michigan-style learning classifier system (LCS) for data mining. In order to handle real-valued inputs effectively, the traditional ternary representation has been replaced by the interval-based representation and the modified XCS has shown to work well. Existing interval-based representations still suffer from a few drawbacks which this paper address. In this paper, we propose an alternative approach called the Min-Percentage representation which produces comparable results to other methods in the literature with the extra advantage of overcoming the drawbacks in these methods. |
