Miguel Melgarejo

Andrea Villate

David Rincon


This paper presents a methodologicalapproach for tuning fuzzy classifiersintended to recognize the Australiansign-language considering twoparticular contexts. We describe thefuzzy classification architecture andthe tuning process based on differentialevolution. The validation resultsshow that it is possible to find a fuzzyclassifier whose classification error isaround 13.0% over a group of wordstaken from several experts for eachinteraction context. This characteristicis relevant as previous works only consideredrecognizing words providedonly by one interpreter.



Auslan, differential evolution, fuzzy classification, pattern recognition, sign language, optimization, TSK Fuzzy Systems

How to Cite
Melgarejo, M., Villate, A., & Rincon, D. (2012). Applying Differential Evolution to Tune Fuzzy Classifiers Intended for Sign-Language Recognition. Ingenieria Y Universidad, 16(2), 397. https://doi.org/10.11144/Javeriana.iyu16-2.adet
Most read articles by the same author(s)