Open Access
Table 1
Summary of machine learning models developed to predict smoke contamination levels and quality traits in wine and berries.
| Product | Type/Algorithm | Inputs | Targets/Outputs | Accuracy | Publication |
|---|---|---|---|---|---|
| Grapevine leaves | Classification Levenberg–Marquardt | 100 Near-infrared absorbance values (1596–2396 nm) | 5 Levels and treatments of smoke taint | 98% | [6] |
| Berries | Classification Levenberg–Marquardt | 100 Near-infrared absorbance values (1596–2396 nm) | 5 Levels and treatments of smoke taint | 97% | [6] |
| Wine | Classification Levenberg–Marquardt | 9 Electronic nose outputs in wine | 5 Levels and treatments of smoke taint in wine | 97% | [10] |
| Wine/Berries | Regression Levenberg–Marquardt | 9 Electronic nose outputs in wine | 20 Glycoconjugates and 10 volatile phenols in berries 1 h after smoking | R = 0.98 | [10] |
| Wine/Berries | Regression Levenberg–Marquardt | 9 Electronic nose outputs in wine | 20 Glycoconjugates and 10 volatile phenols in berries at harvest | R = 0.99 | [10] |
| Wine | Regression Levenberg–Marquardt | 9 Electronic nose outputs in wine | 17 Glycoconjugates and 7 volatile phenols in wine | R = 0.99 | [10] |
| Wine/Berries | Regression Levenberg–Marquardt | 100 Near-infrared absorbance values (1596–2396 nm) in berries | 18 Glycoconjugates and 10 volatile phenols in berries 1 day after smoking | R = 0.98 | [7] |
| Wine/Berries | Regression Levenberg–Marquardt | 100 Near-infrared absorbance values (1596–2396 nm) in berries | 18 Glycoconjugates and 10 volatile phenols in berries at harvest | R = 0.98 | [7] |
| Wine/Berries | Regression Levenberg–Marquardt | 100 Near-infrared absorbance values (1596–2396 nm) in berries | 17 Glycoconjugates and 6 volatile phenols in wine | R = 0.98 | [7] |
| Must and wine | Regression Levenberg–Marquardt | 100 Near-infrared absorbance values (1596–2396 nm) in must | 17 Glycoconjugates and 6 volatile phenols in wine | R = 0.99 | [7] |
| Wine | Regression Levenberg–Marquardt | 100 Near-infrared absorbance values (1596–2396 nm) in wine | 17 Glycoconjugates and 6 volatile phenols in wine | R = 0.99 | [7] |
| Wine | Regression Bayesian Regularization | 9 Electronic nose outputs | Peak area of 8 volatile aromatic compounds | R = 0.99 | [5] |
| Wine | Regression Levenberg–Marquardt | 9 Electronic nose outputs in wine | Liking of 11 sensory attributes, emotion scale, and perceived quality plus the intensity of smoke aroma. | R = 0.98 | [10] |
| Wine | Regression Bayesian Regularization | 9 Electronic nose outputs | Smoke aroma intensity | R = 0.97 | [5] |
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.