Quantification of Spice Mixture Compositions by Electronic Nose: Part I. Experimental Design and Data Analysis Using Neural Networks |
| |
| Authors: | Haoxian Zhang Muratö Balaban José C Principe Kenneth Portier |
| |
| Affiliation: | Author Zhangis with Agricultural and Biological Engineering Dept., Univ. of Florida, Gainesville, Fla.;Author Balaban is with Food Science and Human Nutrition Dept., Univ. of Florida, P.O. Box 110370, Gainesville, FL 32611.;Author Principe is with Computational NeuroEngineering Laboratory, Univ. of Florida, Gainesville, Fla.;Author Portier is with Department of Statistics, Univ. of Florida, Gainesville, Fla. Direct inquiries to author Balaban (E-mail: ). |
| |
| Abstract: | ABSTRACT: A quantitative procedure was developed to predict the composition of ternary ground spice mixtures using an electronic nose. Basil, cinnamon, and garlic were mixed in different compositions and presented to an e-nose. Nineteen training mixtures were used to build predictive models. Model performance was tested using 5 other mixtures. Three neural network structures—multilayer perceptron (MLP), MLP using principal component analysis as a preprocessor (PCA-MLP), and the time-delay neural network (TDNN)—were used for predictive model building. All 3 neural network models predicted the testing mixtures' compositions with a mean square error (MSE) equal or less than 0.0051 (in a fraction domain where sum of fractions = 1). The TDNN provided the smallest MSE. |
| |
| Keywords: | electronic nose mixture neural network quantification spice |
|
|
