Forest Fire Susceptibility Prediction Based on Machine Learning Models with Resampling Algorithms on Remote Sensing Data
This study predicts forest fire susceptibility in Chaloos Rood watershed in Iran using three machine learning (ML) models—multivariate adaptive regression splines (MARS), support vector machine (SVM), and boosted regression tree (BRT). The study utilizes 14 set of fire predictors derived from vegeta...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-11-01
|
| Series: | Remote Sensing |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2072-4292/12/22/3682 |
| _version_ | 1797548364982648832 |
|---|---|
| author | Bahareh Kalantar Naonori Ueda Mohammed O. Idrees Saeid Janizadeh Kourosh Ahmadi Farzin Shabani |
| author_facet | Bahareh Kalantar Naonori Ueda Mohammed O. Idrees Saeid Janizadeh Kourosh Ahmadi Farzin Shabani |
| author_sort | Bahareh Kalantar |
| collection | DOAJ |
| description | This study predicts forest fire susceptibility in Chaloos Rood watershed in Iran using three machine learning (ML) models—multivariate adaptive regression splines (MARS), support vector machine (SVM), and boosted regression tree (BRT). The study utilizes 14 set of fire predictors derived from vegetation indices, climatic variables, environmental factors, and topographical features. To assess the suitability of the models and estimating the variance and bias of estimation, the training dataset obtained from the Natural Resources Directorate of Mazandaran province was subjected to resampling using cross validation (CV), bootstrap, and optimism bootstrap techniques. Using variance inflation factor (VIF), weight indicating the strength of the spatial relationship of the predictors to fire occurrence was assigned to each contributing variable. Subsequently, the models were trained and validated using the receiver operating characteristics (ROC) area under the curve (AUC) curve. Results of the model validation based on the resampling techniques (non, 5- and 10-fold CV, bootstrap and optimism bootstrap) produced AUC values of 0.78, 0.88, 0.90, 0.86 and 0.83 for the MARS model; 0.82, 0.82, 0.89, 0.87, 0.84 for the SVM and 0.87, 0.90, 0.90, 0.90, 0.91 for the BRT model. Across the individual model, the 10-fold CV performed best in MARS and SVM with AUC values of 0.90 and 0.89. Overall, the BRT outperformed the other models in all ramification with highest AUC value of 0.91 using optimism bootstrap resampling algorithm. Generally, the resampling process enhanced the prediction performance of all the models. |
| first_indexed | 2024-03-10T14:57:19Z |
| format | Article |
| id | doaj.art-de7497a02de9414dae2fb6cc16575dc2 |
| institution | Directory Open Access Journal |
| issn | 2072-4292 |
| language | English |
| last_indexed | 2024-03-10T14:57:19Z |
| publishDate | 2020-11-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Remote Sensing |
| spelling | doaj.art-de7497a02de9414dae2fb6cc16575dc22023-11-20T20:28:52ZengMDPI AGRemote Sensing2072-42922020-11-011222368210.3390/rs12223682Forest Fire Susceptibility Prediction Based on Machine Learning Models with Resampling Algorithms on Remote Sensing DataBahareh Kalantar0Naonori Ueda1Mohammed O. Idrees2Saeid Janizadeh3Kourosh Ahmadi4Farzin Shabani5RIKEN Center for Advanced Intelligence Project, Goal-Oriented Technology Research Group, Disaster Resilience Science Team, Tokyo 103-0027, JapanRIKEN Center for Advanced Intelligence Project, Goal-Oriented Technology Research Group, Disaster Resilience Science Team, Tokyo 103-0027, JapanDepartment of Surveying and Geoinformatics, Faculty of Environmental Sciences, University of Ilorin, P.M.B. 1515, 240103 Ilorin, NigeriaDepartment of Watershed Management Engineering, College of Natural Resources, Tarbiat Modares University, Tehran P.O. Box 14115-111, IranDepartment of Forestry, Faculty of Natural Resources and Marine Sciences, Tarbiat Modares University, Tehran 15119-43943, IranDepartment of Biological Sciences, Global Ecology and ARC Centre of Excellence for Australian Biodiversity and Heritage, College of Science and Engineering, Flinders University, GPO Box 2100, Adelaide, SA 5001, AustraliaThis study predicts forest fire susceptibility in Chaloos Rood watershed in Iran using three machine learning (ML) models—multivariate adaptive regression splines (MARS), support vector machine (SVM), and boosted regression tree (BRT). The study utilizes 14 set of fire predictors derived from vegetation indices, climatic variables, environmental factors, and topographical features. To assess the suitability of the models and estimating the variance and bias of estimation, the training dataset obtained from the Natural Resources Directorate of Mazandaran province was subjected to resampling using cross validation (CV), bootstrap, and optimism bootstrap techniques. Using variance inflation factor (VIF), weight indicating the strength of the spatial relationship of the predictors to fire occurrence was assigned to each contributing variable. Subsequently, the models were trained and validated using the receiver operating characteristics (ROC) area under the curve (AUC) curve. Results of the model validation based on the resampling techniques (non, 5- and 10-fold CV, bootstrap and optimism bootstrap) produced AUC values of 0.78, 0.88, 0.90, 0.86 and 0.83 for the MARS model; 0.82, 0.82, 0.89, 0.87, 0.84 for the SVM and 0.87, 0.90, 0.90, 0.90, 0.91 for the BRT model. Across the individual model, the 10-fold CV performed best in MARS and SVM with AUC values of 0.90 and 0.89. Overall, the BRT outperformed the other models in all ramification with highest AUC value of 0.91 using optimism bootstrap resampling algorithm. Generally, the resampling process enhanced the prediction performance of all the models.https://www.mdpi.com/2072-4292/12/22/3682machine learningremote sensingcomputational intelligencebootstrappingcross validation (CV) |
| spellingShingle | Bahareh Kalantar Naonori Ueda Mohammed O. Idrees Saeid Janizadeh Kourosh Ahmadi Farzin Shabani Forest Fire Susceptibility Prediction Based on Machine Learning Models with Resampling Algorithms on Remote Sensing Data Remote Sensing machine learning remote sensing computational intelligence bootstrapping cross validation (CV) |
| title | Forest Fire Susceptibility Prediction Based on Machine Learning Models with Resampling Algorithms on Remote Sensing Data |
| title_full | Forest Fire Susceptibility Prediction Based on Machine Learning Models with Resampling Algorithms on Remote Sensing Data |
| title_fullStr | Forest Fire Susceptibility Prediction Based on Machine Learning Models with Resampling Algorithms on Remote Sensing Data |
| title_full_unstemmed | Forest Fire Susceptibility Prediction Based on Machine Learning Models with Resampling Algorithms on Remote Sensing Data |
| title_short | Forest Fire Susceptibility Prediction Based on Machine Learning Models with Resampling Algorithms on Remote Sensing Data |
| title_sort | forest fire susceptibility prediction based on machine learning models with resampling algorithms on remote sensing data |
| topic | machine learning remote sensing computational intelligence bootstrapping cross validation (CV) |
| url | https://www.mdpi.com/2072-4292/12/22/3682 |
| work_keys_str_mv | AT baharehkalantar forestfiresusceptibilitypredictionbasedonmachinelearningmodelswithresamplingalgorithmsonremotesensingdata AT naonoriueda forestfiresusceptibilitypredictionbasedonmachinelearningmodelswithresamplingalgorithmsonremotesensingdata AT mohammedoidrees forestfiresusceptibilitypredictionbasedonmachinelearningmodelswithresamplingalgorithmsonremotesensingdata AT saeidjanizadeh forestfiresusceptibilitypredictionbasedonmachinelearningmodelswithresamplingalgorithmsonremotesensingdata AT kouroshahmadi forestfiresusceptibilitypredictionbasedonmachinelearningmodelswithresamplingalgorithmsonremotesensingdata AT farzinshabani forestfiresusceptibilitypredictionbasedonmachinelearningmodelswithresamplingalgorithmsonremotesensingdata |