CNCL provides a set of classes for building fuzzy inference engines.
Available are fuzzy sets, fuzzy variables, and a inference engine based
on fuzzy rules. The membership values are normalized ( by default ), but
can be changed to max and min values in different classes. All fuzzy
sets and functions are realized with crisply defined membership values (
type 1 fuzzy sets and functions ).
Currently prod-min inference is used and a center-of-gravity output
defuzzification. This will be extended in a future release, providing
different operators for aggregation, inference, and accumulation. Up to
now two different fuzzy set representations are implemented:
LR-representation and arrays.
For more information about fuzzy logic see:
Zimmermann,. H.-J. [1991]. Fuzzy Set Theory And Its Applications.
Kluwer Accademic Publishers.
Additionally, the FAQ of the newsgroup "comp.ai.fuzzy" should be
recommended in this context.
Go to the first, previous, next, last section, table of contents.