Prediction of Essential Amino Acid Contents in Two Species Cereal Grains Using Artificial Neural Network Approaches

Sarani, F. (2012) Prediction of Essential Amino Acid Contents in Two Species Cereal Grains Using Artificial Neural Network Approaches. Masters thesis, University of Zabol.

[img]
Preview
Text
چکیده انگلیسی1 - cutted.pdf

Download (48kB) | Preview

Abstract

The aim of this study is to evaluate the application of Artificial Neural Networks (ANN) in estimating the amount of essential amino acids with using of the nutrients in wheat and corn. General Regression Neural Network (GRNN), Redial Basis Function (RBF) and Three Layer MLP network are the models used in this study. The training and testing data evaluates neural networks designed in this study. In neural models, we used input variables including crude protein, crude fat, crud fiber, phosphorus and ash, and output variables, including profiles of essential amino acids (methionine, cysteine, leucine, phenylalanine, tryptophan, valine, arginine, lysine, histidine and threonine) to combine these two types of foods. Coefficient of determination was calculated for each nutrient. All three networks were able to acquire the relationship between input and output variables. Results showed that in wheat and corn with input of crude protein+ phosphorus and using of Three Layer MLP network, the coefficient of determination was higher than ever. Coefficient of determination for valine in corn was 0.98 and the coefficient of determination for cysteine in wheat was 0/98. Also in the wheat except methionine, threonine and lysine in the other cases crude protein using general regression neural network and Three Layer MLP network performance was better. The radial basis function in the wheat had not a good performance. Nevertheless, there were opposite results in the corn. Exceptions of tryptophan, arginine and lysine, in the other cases such as redial basis function and whit ash input have a better performance. Using the results of this study recommended that artificial neural networks could be a powerful tool for modeling, forecasting and estimating the nutrient composition of feedstuff used poultry

Item Type: Thesis (Masters)
Uncontrolled Keywords: Artificial Neural Network, Essential Amino Acids, Nutrient, coefficient of determination
Subjects: S Agriculture > SB Plant culture
Depositing User: admin admin1 admin2
Date Deposited: 05 Apr 2016 09:17
Last Modified: 05 Apr 2016 09:17
URI: http://eprints.uoz.ac.ir/id/eprint/257

Actions (login required)

View Item View Item