Preprints
https://doi.org/10.5194/egusphere-2022-1294
https://doi.org/10.5194/egusphere-2022-1294
 
10 Jan 2023
10 Jan 2023
Status: this preprint is open for discussion.

How the recursive feature elimination affects the SVM and RF for wildfire modeling? A mountainous case study area

Ali Rezaei Barzani1, Parham Pahlavani1, Omid Ghorbanzadeh2, and Pedram Ghamisi2,3 Ali Rezaei Barzani et al.
  • 1School of Surveying and Geospatial Engineering, College of Engineering, University of Tehran, Tehran, Iran
  • 2Institute of Advanced Research in Artificial Intelligence (IARAI), Landstraßer Hauptstraße 5, 1030 Vienna, Austria
  • 3Helmholtz-Zentrum Dresden-Rossendorf, Helmholtz Institute Freiberg for Resource Technology, 09599 Freiberg, Germany

Abstract. This study aims to identify key fire factors via recursive feature elimination (RFE) to generate a forest fire susceptibility map (FSM) using support vector machine (SVM), and random forest (RF) models. The fire zones were derived from MODIS satellite imagery from 2012 to 2017. Further validation of these data has been provided by field surveys and reviews of land records in rangelands and forests; a total of 352 fire points were determined in this study. Seventeen factors involving topography, geomorphology, meteorology, hydrology, and anthropology were identified as being effective primary factors in triggering and spreading fires in the selected mountainous case study area. As a first step, the RFE models of the RF, Extra trees, Gradient boosting, and AdaBoost was used to identify important fire factors among all selected primary factors. The SVM and RF models were applied once on all factors and the second on those derived from RFE models as the key factors in FSM. Training and testing data were divided tenfold, and the model's performance was evaluated using cross-validation (CV). Different metrics were used to measure accuracy, including recall, precision, F1, accuracy, the area under the curve (AUC), Matthews correlation coefficient (MCC), and Kappa. The accuracy assessment process shows that the FSM results are further improved by leveraging RFE models to distinguish the key factors and not include unnecessary factors. The greatest improvement is for SVM, with more than 10.97 % and 8.61 % in the accuracy and AUC metrics, respectively.

Ali Rezaei Barzani et al.

Status: open (until 22 Feb 2023)

Comment types: AC – author | RC – referee | CC – community | EC – editor | CEC – chief editor | : Report abuse
  • CC1: 'Comment on egusphere-2022-1294', Carolina Ojeda, 13 Jan 2023 reply

Ali Rezaei Barzani et al.

Ali Rezaei Barzani et al.

Viewed

Total article views: 77 (including HTML, PDF, and XML)
HTML PDF XML Total BibTeX EndNote
50 21 6 77 1 1
  • HTML: 50
  • PDF: 21
  • XML: 6
  • Total: 77
  • BibTeX: 1
  • EndNote: 1
Views and downloads (calculated since 10 Jan 2023)
Cumulative views and downloads (calculated since 10 Jan 2023)

Viewed (geographical distribution)

Total article views: 126 (including HTML, PDF, and XML) Thereof 126 with geography defined and 0 with unknown origin.
Country # Views %
  • 1
1
 
 
 
 
Latest update: 26 Jan 2023
Download
Short summary
• Feature elimination algorithm that combines recursive features with voting for feature selection in forest fire susceptibility assessments. • Using the feature selection method, the primary factors can be reduced, while the evaluation criteria can be improved. • In the assessment of forest fire susceptibility, the random forest method appears to be more accurate than the support vector machine.