TY - JOUR
T1 - Recognizing Odor Mixtures Using Optimized Fuzzy Neural Network Through Genetic Algorithms
AU - Kusumoputro, Benyamin
AU - Arsyad, Teguh P.
N1 - Funding Information:
The authors would like to express their gratitude to Ministry of Research and Technology of Indonesia for their support funding of this research through International Joint Research Program IV under contract number of 9D/PPK/RUTI/KMNRT/II/2005.
Publisher Copyright:
© Fuji Technology Press Ltd. Creative Commons CC BY-ND: This is an Open Access article distributed under the terms of the Creative Commons Attribution-NoDerivatives 4.0 International License (http://creativecommons.org/licenses/by-nd/4.0/).
PY - 2005/5
Y1 - 2005/5
N2 - Recognizing odor mixtures is rather difficult in artificial odor recognition system, especially when the number of sensors is limited. Classification is further hampered if the number of unlearned odor mixtures classes is increased. We developed a fuzzy-neuro multilayer perceptron as a pattern classifier and compared its recognition with that of the Probabilistic Neural Network and Back-propagation Neural Network. To enhance the recognition capability of the system, we then optimized fuzzy-neuro multilayer perceptron topology by deleting its weak weight connections using Genetic Algorithms. Experimental results show that the optimized fuzzy-neuro multilayer perceptron has the highest recognition in 18 classes of two-mixture odors with almost 98.2% when using hardware with 16 sensors, compared to 83.3% when using 8 sensors.
AB - Recognizing odor mixtures is rather difficult in artificial odor recognition system, especially when the number of sensors is limited. Classification is further hampered if the number of unlearned odor mixtures classes is increased. We developed a fuzzy-neuro multilayer perceptron as a pattern classifier and compared its recognition with that of the Probabilistic Neural Network and Back-propagation Neural Network. To enhance the recognition capability of the system, we then optimized fuzzy-neuro multilayer perceptron topology by deleting its weak weight connections using Genetic Algorithms. Experimental results show that the optimized fuzzy-neuro multilayer perceptron has the highest recognition in 18 classes of two-mixture odors with almost 98.2% when using hardware with 16 sensors, compared to 83.3% when using 8 sensors.
KW - fuzzy-neuro system
KW - genetic algorithms
KW - multilayer perceptron
KW - neural structure optimization method
KW - odor recognition system
UR - http://www.scopus.com/inward/record.url?scp=84874678356&partnerID=8YFLogxK
U2 - 10.20965/jaciii.2005.p0290
DO - 10.20965/jaciii.2005.p0290
M3 - Article
AN - SCOPUS:84874678356
SN - 1343-0130
VL - 9
SP - 290
EP - 296
JO - Journal of Advanced Computational Intelligence and Intelligent Informatics
JF - Journal of Advanced Computational Intelligence and Intelligent Informatics
IS - 3
ER -