Predicting land suitability for wheat and barley crops using machine learning techniques
Abstract Ensuring food security to meet the demands of a growing population remains a key challenge, especially for developing countries like Ethiopia. There are various policies and strategies designed by the government and stakeholders to confront the challenge. One of the strategies is using tech...
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Nature Portfolio
2025-05-01
|
| Series: | Scientific Reports |
| Subjects: | |
| Online Access: | https://doi.org/10.1038/s41598-025-99070-0 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1850190657406107648 |
|---|---|
| author | Bikila Abebe Ganati Tilahun Melak Sitote |
| author_facet | Bikila Abebe Ganati Tilahun Melak Sitote |
| author_sort | Bikila Abebe Ganati |
| collection | DOAJ |
| description | Abstract Ensuring food security to meet the demands of a growing population remains a key challenge, especially for developing countries like Ethiopia. There are various policies and strategies designed by the government and stakeholders to confront the challenge. One of the strategies is using technology solutions to increase crop productivity. Precision agriculture using advanced technology has been utilized to increase crop yield. Identifying suitable land for a crop is one of the important factors that will affect the crop’s yield. The existing approach to land suitability identification for a crop is time-consuming, expensive, and inaccurate. In this study, land suitability has been predicted for the two widely grown cereal crops in Ethiopia—wheat and barley—using machine learning techniques. The dataset was obtained from the Engineering Corporation of Oromia (ECO). To make it suitable for modelling, we have pre-processed it. Features have been selected with univariate feature selection (UFS), recursive feature elimination with cross validation (RFECV), and sequential forward selection (SFS). Then, random forest (RF), gradient boosting (GB), and K-nearest neighbour (KNN) were used to predict the land suitability of the two selected crops. To optimize the performance of the models, hyperparameters were tuned with cross-validated randomized searches. The performance of the models has been evaluated using stratified tenfold cross-validation with performance metrics such as accuracy, precision, recall, and F1-score. GB with the SFS has better performance than the other models, with accuracy of 99.41%, precision of 99.37%, recall of 99.34%, and an F1-score of 99.35%. We believe that predicting land suitability accurately using machine learning techniques for the two commonly cultivated cereal crops in Ethiopia will be helpful in increasing the crops’ productivity. The developed model is very accurate. It can be used to develop a decision support system to identify the land suitable for the two crops. |
| format | Article |
| id | doaj-art-4bfbb22785874d9dad27fda4c5a9894d |
| institution | OA Journals |
| issn | 2045-2322 |
| language | English |
| publishDate | 2025-05-01 |
| publisher | Nature Portfolio |
| record_format | Article |
| series | Scientific Reports |
| spelling | doaj-art-4bfbb22785874d9dad27fda4c5a9894d2025-08-20T02:15:12ZengNature PortfolioScientific Reports2045-23222025-05-0115111910.1038/s41598-025-99070-0Predicting land suitability for wheat and barley crops using machine learning techniquesBikila Abebe Ganati0Tilahun Melak Sitote1Department of Computer Science, Faculty of Computing and Informatics, Jimma Institute of Technology, Jimma UniversityDepartment of Software Engineering, College of Engineering, Addis Ababa Science and Technology UniversityAbstract Ensuring food security to meet the demands of a growing population remains a key challenge, especially for developing countries like Ethiopia. There are various policies and strategies designed by the government and stakeholders to confront the challenge. One of the strategies is using technology solutions to increase crop productivity. Precision agriculture using advanced technology has been utilized to increase crop yield. Identifying suitable land for a crop is one of the important factors that will affect the crop’s yield. The existing approach to land suitability identification for a crop is time-consuming, expensive, and inaccurate. In this study, land suitability has been predicted for the two widely grown cereal crops in Ethiopia—wheat and barley—using machine learning techniques. The dataset was obtained from the Engineering Corporation of Oromia (ECO). To make it suitable for modelling, we have pre-processed it. Features have been selected with univariate feature selection (UFS), recursive feature elimination with cross validation (RFECV), and sequential forward selection (SFS). Then, random forest (RF), gradient boosting (GB), and K-nearest neighbour (KNN) were used to predict the land suitability of the two selected crops. To optimize the performance of the models, hyperparameters were tuned with cross-validated randomized searches. The performance of the models has been evaluated using stratified tenfold cross-validation with performance metrics such as accuracy, precision, recall, and F1-score. GB with the SFS has better performance than the other models, with accuracy of 99.41%, precision of 99.37%, recall of 99.34%, and an F1-score of 99.35%. We believe that predicting land suitability accurately using machine learning techniques for the two commonly cultivated cereal crops in Ethiopia will be helpful in increasing the crops’ productivity. The developed model is very accurate. It can be used to develop a decision support system to identify the land suitable for the two crops.https://doi.org/10.1038/s41598-025-99070-0BarleyGradient boostingLand suitabilityMachine learningSequential forward selectionWheat |
| spellingShingle | Bikila Abebe Ganati Tilahun Melak Sitote Predicting land suitability for wheat and barley crops using machine learning techniques Scientific Reports Barley Gradient boosting Land suitability Machine learning Sequential forward selection Wheat |
| title | Predicting land suitability for wheat and barley crops using machine learning techniques |
| title_full | Predicting land suitability for wheat and barley crops using machine learning techniques |
| title_fullStr | Predicting land suitability for wheat and barley crops using machine learning techniques |
| title_full_unstemmed | Predicting land suitability for wheat and barley crops using machine learning techniques |
| title_short | Predicting land suitability for wheat and barley crops using machine learning techniques |
| title_sort | predicting land suitability for wheat and barley crops using machine learning techniques |
| topic | Barley Gradient boosting Land suitability Machine learning Sequential forward selection Wheat |
| url | https://doi.org/10.1038/s41598-025-99070-0 |
| work_keys_str_mv | AT bikilaabebeganati predictinglandsuitabilityforwheatandbarleycropsusingmachinelearningtechniques AT tilahunmelaksitote predictinglandsuitabilityforwheatandbarleycropsusingmachinelearningtechniques |