Premium
Neural Networks vs Principal Component Regression for Prediction of Wheat Flour Loaf Volume in Baking Tests
Author(s) -
HORIMOTO Y.,
DURANCE T.,
NAKAI S.,
LUKOW O.M.
Publication year - 1995
Publication title -
journal of food science
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.772
H-Index - 150
eISSN - 1750-3841
pISSN - 0022-1147
DOI - 10.1111/j.1365-2621.1995.tb09796.x
Subject(s) - principal component analysis , artificial neural network , volume (thermodynamics) , principal component regression , artificial intelligence , centroid , computer science , linear regression , smoothing , component (thermodynamics) , noise (video) , machine learning , regression , regression analysis , statistics , pattern recognition (psychology) , biological system , mathematics , image (mathematics) , physics , quantum mechanics , thermodynamics , biology
Neural networks (NN) provide a simple means of predicting outcomes that depend upon complex, possibly nonlinear, relationships between many variables. A trained neural network was created and used to predict loaf volume of breads made from different wheat cultivars. Although creating the NN required specialized skills and considerable computational time, using the “trained” NN to estimate remix loaf volume, was very rapid and required only basic computer skills. Random Centroid Optimization (RCO) was also employed to choose the best training parameters: learning rate = 0.820, smoothing factor = 0.123, noise = 0.056, number of hidden neurons = 5. NN was more accurate, faster and easier than Principal Component Regression Analysis.