Mapping and Prediction of Community Health Level in The East Java Region Using ANN Backpropagation Method

Abstract

Predicting the level of public health in an area is very important to know the development of public health in order to give consideration to local governments in making policies that can increase the level of health and health services for the community. So we need a system that is able to make predictions and describe the development of public health in the years to come. The data used in the study were the percentage of the population who had health complaints and outpatient treatment for a month, the percentage of households that had access to proper sanitation, the percentage of households that had access to decent drinking water sources, the percentage of poor people, the average length of schooling, infant mortality rate, number of health centers and life expectancy. Backpropagation neural network method is a method that is often used in forecasting or prediction. Where this method can identify activities based on past data. Past data will be studied by neural networks so that they have the ability to make decisions on data that has never been studied. This method is expected to provide a prediction of the level of health by studying data from previous years. The results of this study produced a predictive map of the level of public health in the East Java region in 2019, 2020 and 2021 based on the model obtained from the training results with an epoch of 100 and an MSE value of 0.0173254.