Rule-set is considered as a key element in automatic generation of fuzzy knowledge bases (KB). This paper discusses the influence of the fuzzy-rule tuning on the quality of automatically KBs using a binary coded genetic algorithm (GA). The generated KBs have to satisfy a contradictory paradigm, in term of minimizing the approximation error and the complexity level of the KB. However, the focus is placed on the rule-set tuning of the KBs using specialized reproduction parameters of the GA. The technique is applied on theoretical and practical examples.
|Publication status||Published - 2002|
|Event||Optimization Days 2002 - Montréal, Québec, Canada|
Duration: 1 Jan 2002 → …
|Conference||Optimization Days 2002|
|City||Montréal, Québec, Canada|
|Period||01/01/2002 → …|