Using a comprehensive dataset consisting 1713 samples: 1327 instances of the class “no_fire” and 386 instances of the “fire” class, I used 11 classification models to predict whether or not there is an occurrence of wildfire using the three feature variables: Normalized Difference Vegetation Index (NDVI), Land Surface Temperature (LST) and the Thermal Anomalies (referred to as the BURNED_AREA in the dataset). The aim of this work is to compare the performance of different Machine learning models (with careful hyperparameter selection) for wildfire prediction. The results were obatained for two cases: with balanced class weight and without class weight balance.
More details on the result analysis can be found in my blogpost: https://sites.tufts.edu/olukunleowolabi/2020/03/15/a-comprehensive-classification-model-for-predicting-wildfires-with-uncertainty/