Polynomial regression as a tool for prediction quality of bread baked of wheat flour mixed with flour of cereal extrudates

International Scientific Conference «AGRITECH-2019: Agribusiness, Environmental Engineering and Biotechnologies»

农业科学;生态环境科学;生物科学

Yanova, M.A.^1 ; Oleynikova, E.N.^1 ; Khizhnyak, S.V.^1

Krasnoyarsk State Agrarian University, 90 Mira Avenue, Krasnoyarsk City

660049, Russia^1

关键词: Chemical compositions;    Multiple regression model;    Multivariate polynomial regressions;    Polynomial models;    Polynomial regression;    Polynomial regression models;    Prediction quality;    Titratable acidity;   

Others  :  https://iopscience.iop.org/article/10.1088/1755-1315/315/3/032026/pdf
DOI  :  10.1088/1755-1315/315/3/032026

来源: IOP

【 摘 要 】

Multiple regression models and Multivariate polynomial regression models were used for estimating the relationships between bread quality and chemical composition of wheat flour added with flour of six types of cereal extrudates (wheat extrudate, wheat extrudate added with Fe, covered barley extrudate, naked barley extrudate, covered oat extrudate and naked oat extrudate) in proportion 5, 10, 15, 20 and 25%. Multivariate polynomial regression models were found to be highly effective tool for prediction titratable acidity and porosity and less effective for prediction humidity of bread. Deviation of predicted values of titratable acidity from observed values did not exceed 4.0% of actual value and deviation of predicted values of porosity from observed values did not exceed 0.55% of actual values. The accuracy of models did not depend on the type of extrudate and on the proportion of extrudate in flour. In contrast with polynomial models, the multiple regression models were low effective. Using polynomial regression models two VBA applications for prediction and modeling of titratable acidity and porosity of bread baked from random blends of wheat flour and flour of cereal extrudates were developed.