Abstract
This paper proposes a complexity management methodology for fuzzy systems with feedforward rule bases. The
methodology is based on formal methods for presentation, manipulation and transformation of fuzzy rule bases. First, Boolean
matrices are used for formal presentation of rule bases. Then, binary merging operations are used for formal manipulation of
rule bases. Finally, repetitive merging operations are used for formal transformation of rule bases. The formal methods facilitate
the understanding and modelling of fuzzy systems in terms of interacting subsystems. In particular, the methods reduce the
qualitative complexity in fuzzy systems by improving the transparency of the rule bases.
| Original language | English |
|---|---|
| Article number | 310 |
| Pages (from-to) | 83-95 |
| Number of pages | 13 |
| Journal | International Journal of Knowledge-Based and Intelligent Engineering Systems |
| Volume | 19 |
| DOIs | |
| Publication status | Published - 1 Apr 2015 |
Keywords
- fuzzy systems
- complexity management
- formal methods
- rule bases
- fuzzy networks
- feedforward connections