City Health Prediction Model Using Random Forest Classification Method

City Health Office in Indonesia is creating a health report every year, describing the condition of the city public health. The report is used as the source of determining the city health index. The construction of a city health development index is important to produce an objective formula. In this study, the classification method Random Forest is used to developing a proper model for prediction and analysis of the health index of a city. The goal of this work is to find a prediction model to make a more accurate prediction and reducing errors in dealing with the city health index. The performance of the model is evaluated by using three parameters: Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Squared Error (RMSE). The research shows that the model of Random Forest with a 15 percent data test by using 200 decision trees gives the best results with the value of MAE = 0.108, MSE = 0.035 and RMSE = 0.187, and the Accuracy = 94.6 percent.