Implementation of boosting on the C5.0 algorithm in the health development index data

Abstract

The public health development index is a collection of health indicators that can be easily and directly measured to describe health problems down to the district/city level. The last calculation of the public health development index in Indonesia was carried out in 2018 with the result that not all districts/cities are in a high condition. This research was conducted to classify districts/cities that have a high and low public health development index using the boosted C5.0 algorithm. The results showed that the accuracy of the model increased with the increase in the number of iterations and was constant at the 60th iteration. Error training was also smaller and constant at the 10th iteration. The final accuracy, sensitivity and specificity were obtained respectively 97.09, 96.72, and 97.62.