Using Machine Learning and Feature Selection for Alfalfa Yield Prediction
Predicting alfalfa biomass and crop yield for livestock feed is important to the daily lives of virtually everyone, and many features of data from this domain combined with corresponding weather data can be used to train machine learning models for yield prediction. In this work, we used yield data...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2021-02-01
|
| Series: | AI |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2673-2688/2/1/6 |
| _version_ | 1797396610205876224 |
|---|---|
| author | Christopher D. Whitmire Jonathan M. Vance Hend K. Rasheed Ali Missaoui Khaled M. Rasheed Frederick W. Maier |
| author_facet | Christopher D. Whitmire Jonathan M. Vance Hend K. Rasheed Ali Missaoui Khaled M. Rasheed Frederick W. Maier |
| author_sort | Christopher D. Whitmire |
| collection | DOAJ |
| description | Predicting alfalfa biomass and crop yield for livestock feed is important to the daily lives of virtually everyone, and many features of data from this domain combined with corresponding weather data can be used to train machine learning models for yield prediction. In this work, we used yield data of different alfalfa varieties from multiple years in Kentucky and Georgia, and we compared the impact of different feature selection methods on machine learning (ML) models trained to predict alfalfa yield. Linear regression, regression trees, support vector machines, neural networks, Bayesian regression, and nearest neighbors were all developed with cross validation. The features used included weather data, historical yield data, and the sown date. The feature selection methods that were compared included a correlation-based method, the ReliefF method, and a wrapper method. We found that the best method was the correlation-based method, and the feature set it found consisted of the Julian day of the harvest, the number of days between the sown and harvest dates, cumulative solar radiation since the previous harvest, and cumulative rainfall since the previous harvest. Using these features, the k-nearest neighbor and random forest methods achieved an average R value over 0.95, and average mean absolute error less than 200 lbs./acre. Our top R<sup>2</sup> of 0.90 beats a previous work’s best R<sup>2</sup> of 0.87. Our primary contribution is the demonstration that ML, with feature selection, shows promise in predicting crop yields even on simple datasets with a handful of features, and that reporting accuracies in R and R<sup>2</sup> offers an intuitive way to compare results among various crops. |
| first_indexed | 2024-03-09T00:52:56Z |
| format | Article |
| id | doaj.art-c783f4f52b2943849e33934d3fa1706a |
| institution | Directory Open Access Journal |
| issn | 2673-2688 |
| language | English |
| last_indexed | 2024-03-09T00:52:56Z |
| publishDate | 2021-02-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | AI |
| spelling | doaj.art-c783f4f52b2943849e33934d3fa1706a2023-12-11T17:04:24ZengMDPI AGAI2673-26882021-02-0121718810.3390/ai2010006Using Machine Learning and Feature Selection for Alfalfa Yield PredictionChristopher D. Whitmire0Jonathan M. Vance1Hend K. Rasheed2Ali Missaoui3Khaled M. Rasheed4Frederick W. Maier5Institute for Artificial Intelligence, University of Georgia, 515 Boyd Graduate Studies, 200 D. W. Brooks Drive, Athens, GA 30602, USADepartment of Computer Science, University of Georgia, 415 Boyd Graduate Studies, 200 D. W. Brooks Drive, Athens, GA 30602, USAInstitute for Artificial Intelligence, University of Georgia, 515 Boyd Graduate Studies, 200 D. W. Brooks Drive, Athens, GA 30602, USADepartment of Crop and Soil Sciences, Institute of Plant Breeding Genetics and Genomics, University of Georgia, 4317 Miller Plant Science, Athens, GA 30602, USAInstitute for Artificial Intelligence, University of Georgia, 515 Boyd Graduate Studies, 200 D. W. Brooks Drive, Athens, GA 30602, USAInstitute for Artificial Intelligence, University of Georgia, 515 Boyd Graduate Studies, 200 D. W. Brooks Drive, Athens, GA 30602, USAPredicting alfalfa biomass and crop yield for livestock feed is important to the daily lives of virtually everyone, and many features of data from this domain combined with corresponding weather data can be used to train machine learning models for yield prediction. In this work, we used yield data of different alfalfa varieties from multiple years in Kentucky and Georgia, and we compared the impact of different feature selection methods on machine learning (ML) models trained to predict alfalfa yield. Linear regression, regression trees, support vector machines, neural networks, Bayesian regression, and nearest neighbors were all developed with cross validation. The features used included weather data, historical yield data, and the sown date. The feature selection methods that were compared included a correlation-based method, the ReliefF method, and a wrapper method. We found that the best method was the correlation-based method, and the feature set it found consisted of the Julian day of the harvest, the number of days between the sown and harvest dates, cumulative solar radiation since the previous harvest, and cumulative rainfall since the previous harvest. Using these features, the k-nearest neighbor and random forest methods achieved an average R value over 0.95, and average mean absolute error less than 200 lbs./acre. Our top R<sup>2</sup> of 0.90 beats a previous work’s best R<sup>2</sup> of 0.87. Our primary contribution is the demonstration that ML, with feature selection, shows promise in predicting crop yields even on simple datasets with a handful of features, and that reporting accuracies in R and R<sup>2</sup> offers an intuitive way to compare results among various crops.https://www.mdpi.com/2673-2688/2/1/6alfalfacross validationfeature selectionmachine learningregressionyield prediction |
| spellingShingle | Christopher D. Whitmire Jonathan M. Vance Hend K. Rasheed Ali Missaoui Khaled M. Rasheed Frederick W. Maier Using Machine Learning and Feature Selection for Alfalfa Yield Prediction AI alfalfa cross validation feature selection machine learning regression yield prediction |
| title | Using Machine Learning and Feature Selection for Alfalfa Yield Prediction |
| title_full | Using Machine Learning and Feature Selection for Alfalfa Yield Prediction |
| title_fullStr | Using Machine Learning and Feature Selection for Alfalfa Yield Prediction |
| title_full_unstemmed | Using Machine Learning and Feature Selection for Alfalfa Yield Prediction |
| title_short | Using Machine Learning and Feature Selection for Alfalfa Yield Prediction |
| title_sort | using machine learning and feature selection for alfalfa yield prediction |
| topic | alfalfa cross validation feature selection machine learning regression yield prediction |
| url | https://www.mdpi.com/2673-2688/2/1/6 |
| work_keys_str_mv | AT christopherdwhitmire usingmachinelearningandfeatureselectionforalfalfayieldprediction AT jonathanmvance usingmachinelearningandfeatureselectionforalfalfayieldprediction AT hendkrasheed usingmachinelearningandfeatureselectionforalfalfayieldprediction AT alimissaoui usingmachinelearningandfeatureselectionforalfalfayieldprediction AT khaledmrasheed usingmachinelearningandfeatureselectionforalfalfayieldprediction AT frederickwmaier usingmachinelearningandfeatureselectionforalfalfayieldprediction |