Advertisement

Final Remarks

Chapter
  • 318 Downloads
Part of the Studies in Computational Intelligence book series (SCI, volume 163)

Abstract

In this monograph, we studied inhibitory decision and association rules. We showed that using inhibitory rules one can describe more knowledge encoded in information and decision systems than in the case of deterministic (standard) rules.

Unfortunately, for almost all k-valued information systems with the polynomial number of objects in the number of attributes the number of minimal (irreducible) inhibitory association rules is not polynomial in the number of attributes. In some sense analogous situation is with minimal inhibitory decision rules.

In such a situation, we can either use some heuristics for generating of relatively small sets of “important” inhibitory rules, or use lazy classification algorithms which in polynomial time can find an information about the whole set of true and realizable inhibitory rules for a given information or decision system.

Copyright information

© Springer-Verlag Berlin Heidelberg 2009

Authors and Affiliations

  1. 1.University of RzeszowPoland
  2. 2.University of Silesia, SosnowiecPoland
  3. 3.Warsaw University, WarsawPoland

Personalised recommendations