Algorithms for predicting fire: a comparative study for application in sugar cane plantation

The occurrence of fire in sugarcane fields is a problem that has plagued several regions in Brazil for a long time. Due to the concern regarding its tragic consequences, it is necessary to take action to avoid or minimize its occurrence. With the development of Artificial Intelligence (AI) and the growing use of Machine Learning (ML), there is an opportunity to use technology in order to preview an imminent fire. In this context, the objective of this work is to implement predictive algorithms using the Python language to compare and define what best applies to fire prediction in sugarcane fields. The data necessary for this purpose were provided by a sugarcane company located in the State of São Paulo, Brazil. Among the four algorithms analysed, which were the Support Vector Machine (SVM), Naive Bayes, Random Forest, and XGBoost, the SVM model showed better performance with the metrics used.