Assessment of Various Machine Learning Models for Peach Maturity Prediction Using Non-Destructive Sensor Data
To date, many machine learning models have been used for peach maturity prediction using non-destructive data, but no performance comparison of the models on these datasets has been conducted. In this study, eight machine learning models were trained on a dataset containing data from 180 ‘Suncrest’...
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2022-08-01
|
| Series: | Sensors |
| Subjects: | |
| Online Access: | https://www.mdpi.com/1424-8220/22/15/5791 |
| _version_ | 1797440647927431168 |
|---|---|
| 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 | To date, many machine learning models have been used for peach maturity prediction using non-destructive data, but no performance comparison of the models on these datasets has been conducted. In this study, eight machine learning models were trained on a dataset containing data from 180 ‘Suncrest’ peaches. Before the models were trained, the dataset was subjected to dimensionality reduction using the least absolute shrinkage and selection operator (LASSO) regularization, and 8 input variables (out of 29) were chosen. At the same time, a subgroup consisting of the peach ground color measurements was singled out by dividing the set of variables into three subgroups and by using group LASSO regularization. This type of variable subgroup selection provided valuable information on the contribution of specific groups of peach traits to the maturity prediction. The area under the receiver operating characteristic curve (AUC) values of the selected models were compared, and the artificial neural network (ANN) model achieved the best performance, with an average AUC of 0.782. The second-best machine learning model was linear discriminant analysis with an AUC of 0.766, followed by logistic regression, gradient boosting machine, random forest, support vector machines, a classification and regression trees model, and k-nearest neighbors. Although the primary parameter used to determine the performance of the model was AUC, accuracy, F1 score, and kappa served as control parameters and ultimately confirmed the obtained results. By outperforming other models, ANN proved to be the most accurate model for peach maturity prediction on the given dataset. |
| first_indexed | 2024-03-09T12:12:17Z |
| format | Article |
| id | doaj.art-f51827de7b3c4306bdd806739dcc82d4 |
| institution | Directory Open Access Journal |
| issn | 1424-8220 |
| language | English |
| last_indexed | 2024-03-09T12:12:17Z |
| publishDate | 2022-08-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Sensors |
| spelling | doaj.art-f51827de7b3c4306bdd806739dcc82d42023-11-30T22:51:50ZengMDPI AGSensors1424-82202022-08-012215579110.3390/s22155791Assessment of Various Machine Learning Models for Peach Maturity Prediction Using Non-Destructive Sensor DataDejan Ljubobratović0Marko Vuković1Marija Brkić Bakarić2Tomislav Jemrić3Maja Matetić4Faculty of Informatics and Digital Technologies, University of Rijeka, Radmile Matejčić 2, 51000 Rijeka, CroatiaDivision of Horticulture and Landscape Architecture, Department of Pomology, Svetošimunska cesta 25, University of Zagreb Faculty of Agriculture, 10000 Zagreb, CroatiaFaculty of Informatics and Digital Technologies, University of Rijeka, Radmile Matejčić 2, 51000 Rijeka, CroatiaDivision of Horticulture and Landscape Architecture, Department of Pomology, Svetošimunska cesta 25, University of Zagreb Faculty of Agriculture, 10000 Zagreb, CroatiaFaculty of Informatics and Digital Technologies, University of Rijeka, Radmile Matejčić 2, 51000 Rijeka, CroatiaTo date, many machine learning models have been used for peach maturity prediction using non-destructive data, but no performance comparison of the models on these datasets has been conducted. In this study, eight machine learning models were trained on a dataset containing data from 180 ‘Suncrest’ peaches. Before the models were trained, the dataset was subjected to dimensionality reduction using the least absolute shrinkage and selection operator (LASSO) regularization, and 8 input variables (out of 29) were chosen. At the same time, a subgroup consisting of the peach ground color measurements was singled out by dividing the set of variables into three subgroups and by using group LASSO regularization. This type of variable subgroup selection provided valuable information on the contribution of specific groups of peach traits to the maturity prediction. The area under the receiver operating characteristic curve (AUC) values of the selected models were compared, and the artificial neural network (ANN) model achieved the best performance, with an average AUC of 0.782. The second-best machine learning model was linear discriminant analysis with an AUC of 0.766, followed by logistic regression, gradient boosting machine, random forest, support vector machines, a classification and regression trees model, and k-nearest neighbors. Although the primary parameter used to determine the performance of the model was AUC, accuracy, F1 score, and kappa served as control parameters and ultimately confirmed the obtained results. By outperforming other models, ANN proved to be the most accurate model for peach maturity prediction on the given dataset.https://www.mdpi.com/1424-8220/22/15/5791machine learningAUCpeach maturity predictionartificial neural networksfruit qualitynon-destructive measurements |
| spellingShingle | Dejan Ljubobratović Marko Vuković Marija Brkić Bakarić Tomislav Jemrić Maja Matetić Assessment of Various Machine Learning Models for Peach Maturity Prediction Using Non-Destructive Sensor Data Sensors machine learning AUC peach maturity prediction artificial neural networks fruit quality non-destructive measurements |
| title | Assessment of Various Machine Learning Models for Peach Maturity Prediction Using Non-Destructive Sensor Data |
| title_full | Assessment of Various Machine Learning Models for Peach Maturity Prediction Using Non-Destructive Sensor Data |
| title_fullStr | Assessment of Various Machine Learning Models for Peach Maturity Prediction Using Non-Destructive Sensor Data |
| title_full_unstemmed | Assessment of Various Machine Learning Models for Peach Maturity Prediction Using Non-Destructive Sensor Data |
| title_short | Assessment of Various Machine Learning Models for Peach Maturity Prediction Using Non-Destructive Sensor Data |
| title_sort | assessment of various machine learning models for peach maturity prediction using non destructive sensor data |
| topic | machine learning AUC peach maturity prediction artificial neural networks fruit quality non-destructive measurements |
| url | https://www.mdpi.com/1424-8220/22/15/5791 |
| work_keys_str_mv | AT dejanljubobratovic assessmentofvariousmachinelearningmodelsforpeachmaturitypredictionusingnondestructivesensordata AT markovukovic assessmentofvariousmachinelearningmodelsforpeachmaturitypredictionusingnondestructivesensordata AT marijabrkicbakaric assessmentofvariousmachinelearningmodelsforpeachmaturitypredictionusingnondestructivesensordata AT tomislavjemric assessmentofvariousmachinelearningmodelsforpeachmaturitypredictionusingnondestructivesensordata AT majamatetic assessmentofvariousmachinelearningmodelsforpeachmaturitypredictionusingnondestructivesensordata |