Cui, Chenhao;
Caporaso, Nicola;
Chen, Jiawei;
Fearn, Tom;
(2023)
Farinograph characteristics of wheat flour predicted by near infrared spectroscopy with an ensemble modelling method.
Journal of Food Engineering
, 359
, Article 111689. 10.1016/j.jfoodeng.2023.111689.
Preview |
Text
Farinograph_JFE_Pre-proof.pdf - Accepted Version Download (3MB) | Preview |
Abstract
The quality and end use of wheat flour is linked to its farinograph characteristics, namely water absorption, dough development time, dough stability, and degree of softening. Near infrared (NIR) spectroscopy has been previously used to predict these properties, but the reported results are based on calibration samples that represent relatively specialized populations. In this study, 1028 samples of wheat flour (including both hard and soft wheats, with protein range ∼7–19%) were collected from a wide range of countries and continents, proposing an ensemble method of classification and regression models to improve the prediction performance of farinograph properties by NIR spectroscopy. The outputs of a Gaussian process regression (GPR) were used to assign samples to two groups, for which two local partial least squares regression (PLSR) models predicted farinograph properties. Our modelling method significantly improved the prediction performance of farinograph characteristics. The results for predicting water absorption, dough development time, dough stability, and degree of softening were satisfactory and useable for a rapid estimation of wheat flour quality. In comparison to a global PLSR model, modelling parameter estimators indicated superior performance of this novel method. This improved prediction performance can provide significant benefits to the cereal and baking industry in terms of in-line rapid analysis and accurate prediction of flour properties and quality.
| Type: | Article |
|---|---|
| Title: | Farinograph characteristics of wheat flour predicted by near infrared spectroscopy with an ensemble modelling method |
| Open access status: | An open access version is available from UCL Discovery |
| DOI: | 10.1016/j.jfoodeng.2023.111689 |
| Publisher version: | https://doi.org/10.1016/j.jfoodeng.2023.111689 |
| Language: | English |
| Additional information: | This version is the author-accepted manuscript. For information on re-use, please refer to the publisher’s terms and conditions. |
| Keywords: | Farinograph characteristics, Wheat flour properties, NIR spectroscopy, Partial least square regression, Gaussian process regression, Chemometrics |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Maths and Physical Sciences > Dept of Statistical Science |
| URI: | https://discovery-pp.ucl.ac.uk/id/eprint/10178052 |
Archive Staff Only
 |
View Item |
