Key points are not available for this paper at this time.
This paper presents the results of modeling the risk of forest fires in the west of Nghe An Province (north-central Vietnam) using remote sensing and GIS data. The nine factors influencing the risk of forest fires, including vegetation cover (NDVI vegetation index), surface evapotranspiration, elevation (DEM), slope (slope), aspect, wind speed, ground surface temperature, average monthly precipitation and population density are used to build a forest fire risk mapping model based on machine learning methods, including Random Forest (RF), Suppor Vector Machine (SVM), and Classification and Regression Trees (CART). Various parameters are tested in the RF, SVM, CART algorithms to select the algorithm with the highest accuracy in forest fire risk prediction. The obtained results show that the RF algorithm with the value of the numberOfTrees parameter equal to 100 has the highest accuracy in predicting the risk of forest fires in the study area, expressed through the location of the distribution of forest fire points, as well as the AUC value on the ROC curve. The results obtained in the study can be effectively used for monitoring and early warning of forest fire danger in settlements, helping to reduce damage from forest fires.
Doan et al. (Thu,) studied this question.