Predicting Forest Fire with Linear Regression and Random Forest

Abstract

Forest fire is a global problem and it bothers a lot of countries. It not only threatens human life, but also causes serious damage to the environment and great economic losses. Therefore, forest fire is particularly important for forest fire prevention. and economic loss. However there many methods can be used to predict the forest fires then the forest fires could be stopped even before it happened. So, in the research will be talking about two methods in analyzing forest fire data set in order to predict the forest fires, they are linear regression and random forest. And before analyzing the data, the data will be pre-processed by PCA in order to get a more accurate result. Finally compare the result from linear regression and random forest to compare which method has a higher accuracy and better performance. And from the result and the comparison linear regression showed to be perform better in predicting forest fire.