Regression Models Analysis in Predicting the Impact of Population Growth on Flood Intensity in Jakarta

Abstract

The population in Jakarta has been in an increasing trend since the early 1950’s. This inclination trend has adverse effect towards the intensity of the flood that occur in Jakarta. It is essential to be able to predict the impact of the increasing trend of population growth on flood intensity and prepare for possible outcomes in future. By using machine learning methods, this paper aims to determine the most suitable regression methods in building regression models to predict the impact of population growth on flood intensity in Jakarta. We build and compare the performances of four regression models: linear regression, polynomial regression, ridge regression, and decision tree regression. For the dataset, we collected population and flood data from 2013-2020. We duplicate them, one is split into train and test set, and the other is not. Through our experiments, we found out that the best regression methods are polynomial regression and decision tree regression.