This research aims to develop a prediction model for forest fires that occur by utilizing five types of machine learning algorithms, namely Decision Tree, K-Nearest Neighbors (KNN), Random Forest, Naïve Bayes (Kernel), and Rule Induction. The data used in this research was taken from www.kaggle.com. By using data pre-processing techniques such as missing value imputation, data normalization, and feature selection techniques, to ensure the quality of the data used in the prediction model. The research results show that each algorithm has different performance in predicting forest fires that occur each month, with some algorithms showing higher levels of accuracy and precision. Further analysis discusses the advantages and disadvantages of each algorithm as well as the practical implications of implementing them in the environment.
Ruth Meylina Manik (Thu,) studied this question.