Comprehensive Classification and Regression Modeling of Wine Samples Using <sup>1</sup>H NMR Spectra
Recently, <sup>1</sup>H NMR (nuclear magnetic resonance) spectroscopy was presented as a viable option for the quality assurance of foods and beverages, such as wine products. Here, a complex chemometric analysis of red and white wine samples was carried out based on their <sup>1&l...
Main Authors: | , , , , |
---|---|
Format: | Article |
Language: | English |
Published: |
MDPI AG
2020-12-01
|
Series: | Foods |
Subjects: | |
Online Access: | https://www.mdpi.com/2304-8158/10/1/64 |
_version_ | 1797543093668413440 |
---|---|
author | Gábor Barátossy Mária Berinkeiné Donkó Helga Csikorné Vásárhelyi Károly Héberger Anita Rácz |
author_facet | Gábor Barátossy Mária Berinkeiné Donkó Helga Csikorné Vásárhelyi Károly Héberger Anita Rácz |
author_sort | Gábor Barátossy |
collection | DOAJ |
description | Recently, <sup>1</sup>H NMR (nuclear magnetic resonance) spectroscopy was presented as a viable option for the quality assurance of foods and beverages, such as wine products. Here, a complex chemometric analysis of red and white wine samples was carried out based on their <sup>1</sup>H NMR spectra. Extreme gradient boosting (XGBoost) machine learning algorithm was applied for the wine variety classification with an iterative double cross-validation loop, developed during the present work. In the case of red wines, Cabernet Franc, Merlot and Blue Frankish samples were successfully classified. Three very common white wine varieties were selected and classified: Chardonnay, Sauvignon Blanc and Riesling. The models were robust and were validated against overfitting with iterative randomization tests. Moreover, four novel partial least-squares (PLS) regression models were constructed to predict the major quantitative parameters of the wines: density, total alcohol, total sugar and total SO<sub>2</sub> concentrations. All the models performed successfully, with <i>R</i><sup>2</sup> values above 0.80 in almost every case, providing additional information about the wine samples for the quality control of the products. <sup>1</sup>H NMR spectra combined with chemometric modeling can be a good and reliable candidate for the replacement of the time-consuming traditional standards, not just in wine analysis, but also in other aspects of food science. |
first_indexed | 2024-03-10T13:40:45Z |
format | Article |
id | doaj.art-889f2c93ba17454ea8fd04e701874256 |
institution | Directory Open Access Journal |
issn | 2304-8158 |
language | English |
last_indexed | 2024-03-10T13:40:45Z |
publishDate | 2020-12-01 |
publisher | MDPI AG |
record_format | Article |
series | Foods |
spelling | doaj.art-889f2c93ba17454ea8fd04e7018742562023-11-21T03:04:30ZengMDPI AGFoods2304-81582020-12-011016410.3390/foods10010064Comprehensive Classification and Regression Modeling of Wine Samples Using <sup>1</sup>H NMR SpectraGábor Barátossy0Mária Berinkeiné Donkó1Helga Csikorné Vásárhelyi2Károly Héberger3Anita Rácz4National Food Chain Safety Office, Directorate of Oenology and Alcoholic Beverages, Budaörsi út 141-145, H-1118 Budapest, HungaryNational Food Chain Safety Office, Directorate of Oenology and Alcoholic Beverages, Budaörsi út 141-145, H-1118 Budapest, HungaryNational Food Chain Safety Office, Directorate of Oenology and Alcoholic Beverages, Budaörsi út 141-145, H-1118 Budapest, HungaryDepartment of Plasma Chemistry, Institute of Materials and Environmental Chemistry, Research Centre for Natural Sciences, Magyar Tudósok krt. 2, H-1117 Budapest, HungaryDepartment of Plasma Chemistry, Institute of Materials and Environmental Chemistry, Research Centre for Natural Sciences, Magyar Tudósok krt. 2, H-1117 Budapest, HungaryRecently, <sup>1</sup>H NMR (nuclear magnetic resonance) spectroscopy was presented as a viable option for the quality assurance of foods and beverages, such as wine products. Here, a complex chemometric analysis of red and white wine samples was carried out based on their <sup>1</sup>H NMR spectra. Extreme gradient boosting (XGBoost) machine learning algorithm was applied for the wine variety classification with an iterative double cross-validation loop, developed during the present work. In the case of red wines, Cabernet Franc, Merlot and Blue Frankish samples were successfully classified. Three very common white wine varieties were selected and classified: Chardonnay, Sauvignon Blanc and Riesling. The models were robust and were validated against overfitting with iterative randomization tests. Moreover, four novel partial least-squares (PLS) regression models were constructed to predict the major quantitative parameters of the wines: density, total alcohol, total sugar and total SO<sub>2</sub> concentrations. All the models performed successfully, with <i>R</i><sup>2</sup> values above 0.80 in almost every case, providing additional information about the wine samples for the quality control of the products. <sup>1</sup>H NMR spectra combined with chemometric modeling can be a good and reliable candidate for the replacement of the time-consuming traditional standards, not just in wine analysis, but also in other aspects of food science.https://www.mdpi.com/2304-8158/10/1/64winemachine learningspectroscopycross-validationmetabolomics |
spellingShingle | Gábor Barátossy Mária Berinkeiné Donkó Helga Csikorné Vásárhelyi Károly Héberger Anita Rácz Comprehensive Classification and Regression Modeling of Wine Samples Using <sup>1</sup>H NMR Spectra Foods wine machine learning spectroscopy cross-validation metabolomics |
title | Comprehensive Classification and Regression Modeling of Wine Samples Using <sup>1</sup>H NMR Spectra |
title_full | Comprehensive Classification and Regression Modeling of Wine Samples Using <sup>1</sup>H NMR Spectra |
title_fullStr | Comprehensive Classification and Regression Modeling of Wine Samples Using <sup>1</sup>H NMR Spectra |
title_full_unstemmed | Comprehensive Classification and Regression Modeling of Wine Samples Using <sup>1</sup>H NMR Spectra |
title_short | Comprehensive Classification and Regression Modeling of Wine Samples Using <sup>1</sup>H NMR Spectra |
title_sort | comprehensive classification and regression modeling of wine samples using sup 1 sup h nmr spectra |
topic | wine machine learning spectroscopy cross-validation metabolomics |
url | https://www.mdpi.com/2304-8158/10/1/64 |
work_keys_str_mv | AT gaborbaratossy comprehensiveclassificationandregressionmodelingofwinesamplesusingsup1suphnmrspectra AT mariaberinkeinedonko comprehensiveclassificationandregressionmodelingofwinesamplesusingsup1suphnmrspectra AT helgacsikornevasarhelyi comprehensiveclassificationandregressionmodelingofwinesamplesusingsup1suphnmrspectra AT karolyheberger comprehensiveclassificationandregressionmodelingofwinesamplesusingsup1suphnmrspectra AT anitaracz comprehensiveclassificationandregressionmodelingofwinesamplesusingsup1suphnmrspectra |