Utilization of Explainable Machine Learning Algorithms for Determination of Important Features in ‘Suncrest’ Peach Maturity Prediction
Peaches (<i>Prunus persica</i> (L.) Batsch) are a popular fruit in Europe and Croatia. Maturity at harvest has a crucial influence on peach fruit quality, storage life, and consequently consumer acceptance. The main goal of this study is to develop a machine learning model that will dete...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-12-01
|
| Series: | Electronics |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2079-9292/10/24/3115 |
| _version_ | 1797505196081807360 |
|---|---|
| author | Dejan Ljubobratović Marko Vuković Marija Brkić Bakarić Tomislav Jemrić Maja Matetić |
| author_facet | Dejan Ljubobratović Marko Vuković Marija Brkić Bakarić Tomislav Jemrić Maja Matetić |
| author_sort | Dejan Ljubobratović |
| collection | DOAJ |
| description | Peaches (<i>Prunus persica</i> (L.) Batsch) are a popular fruit in Europe and Croatia. Maturity at harvest has a crucial influence on peach fruit quality, storage life, and consequently consumer acceptance. The main goal of this study is to develop a machine learning model that will detect the most important features for predicting peach maturity by first training models and then using the importance ratings of these models to detect nonlinear (and linear) relationships. Thus, the most important peach features at a given stage of its ripening could be revealed. To date, this method has not been used for this purpose, and at the same time, it has the potential to be applied to other similar peach varieties. A total of 33 fruit features are measured on the harvested peaches, and three imbalanced datasets are created using firmness thresholds of 1.84, 3.57, and 4.59 kg·cm<sup>−2</sup>. These datasets are balanced using the SMOTE and ROSE techniques, and the Random Forest machine learning model is trained on them. Permutation Feature Importance (PFI), Variable Importance (VI), and LIME interpretability methods are used to detect variables that most influence predictions in the given machine learning models. PFI shows that the <i>h°</i> and <i>a</i>* ground color parameters, COL ground color index, SSC/TA, and TA inner quality parameters are among the top ten most contributing variables in all three models. Meanwhile, VI shows that this is the case for the <i>a</i>* ground color parameter, COL and CCL ground color indexes, and the SSC/TA inner quality parameter. The fruit flesh ratio is highly positioned (among the top three according to PFI) in two models, but it is not even among the top ten in the third. |
| first_indexed | 2024-03-10T04:15:09Z |
| format | Article |
| id | doaj.art-8f815f40883a4e6eafd48ed5c7db8a14 |
| institution | Directory Open Access Journal |
| issn | 2079-9292 |
| language | English |
| last_indexed | 2024-03-10T04:15:09Z |
| publishDate | 2021-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Electronics |
| spelling | doaj.art-8f815f40883a4e6eafd48ed5c7db8a142023-11-23T08:02:19ZengMDPI AGElectronics2079-92922021-12-011024311510.3390/electronics10243115Utilization of Explainable Machine Learning Algorithms for Determination of Important Features in ‘Suncrest’ Peach Maturity PredictionDejan Ljubobratović0Marko Vuković1Marija Brkić Bakarić2Tomislav Jemrić3Maja Matetić4Department of Informatics, University of Rijeka, Radmile Matejčić 2, 51000 Rijeka, CroatiaFaculty of Agriculture, Unit of Horticulture and Landscape Architecture, Department of Pomology, University of Zagreb, Svetošimunska c. 25, 10000 Zagreb, CroatiaDepartment of Informatics, University of Rijeka, Radmile Matejčić 2, 51000 Rijeka, CroatiaFaculty of Agriculture, Unit of Horticulture and Landscape Architecture, Department of Pomology, University of Zagreb, Svetošimunska c. 25, 10000 Zagreb, CroatiaDepartment of Informatics, University of Rijeka, Radmile Matejčić 2, 51000 Rijeka, CroatiaPeaches (<i>Prunus persica</i> (L.) Batsch) are a popular fruit in Europe and Croatia. Maturity at harvest has a crucial influence on peach fruit quality, storage life, and consequently consumer acceptance. The main goal of this study is to develop a machine learning model that will detect the most important features for predicting peach maturity by first training models and then using the importance ratings of these models to detect nonlinear (and linear) relationships. Thus, the most important peach features at a given stage of its ripening could be revealed. To date, this method has not been used for this purpose, and at the same time, it has the potential to be applied to other similar peach varieties. A total of 33 fruit features are measured on the harvested peaches, and three imbalanced datasets are created using firmness thresholds of 1.84, 3.57, and 4.59 kg·cm<sup>−2</sup>. These datasets are balanced using the SMOTE and ROSE techniques, and the Random Forest machine learning model is trained on them. Permutation Feature Importance (PFI), Variable Importance (VI), and LIME interpretability methods are used to detect variables that most influence predictions in the given machine learning models. PFI shows that the <i>h°</i> and <i>a</i>* ground color parameters, COL ground color index, SSC/TA, and TA inner quality parameters are among the top ten most contributing variables in all three models. Meanwhile, VI shows that this is the case for the <i>a</i>* ground color parameter, COL and CCL ground color indexes, and the SSC/TA inner quality parameter. The fruit flesh ratio is highly positioned (among the top three according to PFI) in two models, but it is not even among the top ten in the third.https://www.mdpi.com/2079-9292/10/24/3115machine learningimbalanced datasetspeach maturityvariable importanceinterpretable machine learningrandom forest |
| spellingShingle | Dejan Ljubobratović Marko Vuković Marija Brkić Bakarić Tomislav Jemrić Maja Matetić Utilization of Explainable Machine Learning Algorithms for Determination of Important Features in ‘Suncrest’ Peach Maturity Prediction Electronics machine learning imbalanced datasets peach maturity variable importance interpretable machine learning random forest |
| title | Utilization of Explainable Machine Learning Algorithms for Determination of Important Features in ‘Suncrest’ Peach Maturity Prediction |
| title_full | Utilization of Explainable Machine Learning Algorithms for Determination of Important Features in ‘Suncrest’ Peach Maturity Prediction |
| title_fullStr | Utilization of Explainable Machine Learning Algorithms for Determination of Important Features in ‘Suncrest’ Peach Maturity Prediction |
| title_full_unstemmed | Utilization of Explainable Machine Learning Algorithms for Determination of Important Features in ‘Suncrest’ Peach Maturity Prediction |
| title_short | Utilization of Explainable Machine Learning Algorithms for Determination of Important Features in ‘Suncrest’ Peach Maturity Prediction |
| title_sort | utilization of explainable machine learning algorithms for determination of important features in suncrest peach maturity prediction |
| topic | machine learning imbalanced datasets peach maturity variable importance interpretable machine learning random forest |
| url | https://www.mdpi.com/2079-9292/10/24/3115 |
| work_keys_str_mv | AT dejanljubobratovic utilizationofexplainablemachinelearningalgorithmsfordeterminationofimportantfeaturesinsuncrestpeachmaturityprediction AT markovukovic utilizationofexplainablemachinelearningalgorithmsfordeterminationofimportantfeaturesinsuncrestpeachmaturityprediction AT marijabrkicbakaric utilizationofexplainablemachinelearningalgorithmsfordeterminationofimportantfeaturesinsuncrestpeachmaturityprediction AT tomislavjemric utilizationofexplainablemachinelearningalgorithmsfordeterminationofimportantfeaturesinsuncrestpeachmaturityprediction AT majamatetic utilizationofexplainablemachinelearningalgorithmsfordeterminationofimportantfeaturesinsuncrestpeachmaturityprediction |