Prediction of Pneumonia Patient Proportions Using Kriging and Inverse Distance Weighted Methods

Abstract

Pneumonia is a respiratory tract disease caused by bacteria, viruses, or fungi. In Indonesia, pneumonia is one of the diseases with the second highest number of cases after malaria from 2007 to 2015. Mapping of regions with pneumonia prevalence needs to be done so that the government can pay more attention to areas with high pneumonia rates, and the public can be more aware of areas that are prone to pneumonia. Spatial methods that can be used to predict the distribution of pneumonia patients are ordinary kriging and inverse distance weighted. The research results showed that the best method chosen for predicting the proportion of pneumonia patients in Sumatra, Java, Kalimantan, and Sulawesi islands is the ordinary kriging method with its theoretical semivariogram model being exponential, and the obtained RMSE value is 0.0178. The average proportion of pneumonia patients in Sumatra Island is 0.01655, with the highest proportion value of 0.02504 in Simalungun District, North Sumatra. Then, the average proportion of pneumonia patients in Java Island is 0.01672, with the highest proportion value of 0.02443 in Garut District, West Java. The proportion of pneumonia patients in Kalimantan Island is 0.01657, with the highest proportion value of 0.01835 in Balikpapan District, East Kalimantan. The average proportion of pneumonia patients in Sulawesi Island is 0.0166, with the highest proportion value of 0.02397 in South Minahasa District, North Sulawesi.