Application of backpropagation artificial neural network to predict human development index of Maluku Province

Abstract

The Human Development Index (HDI)) is a comparative measurement of longevity and healthy living, education, and a decent standard of living, for all countries throughout the world. HDI is used to classify whether a country is a developed country, a developing country or an underdeveloped country and also to measure the effect of economic policies on quality of life. In Indonesia HDI is one of the important indicators in measuring success in efforts to build the quality of human life (community/population), determine the rank or level of development of an area and as a allocator for determining general allocation funds. However, in reality the calculation and publication time of HDI by the Central Bureau of Statistics takes quite a long time. So this research aims to predict the HDI value of Maluku Province for the next 5 years using the Artificial Neural Network (ANN) Backpropagation method. And also, design or build an application system with the Graphic User Interface (GUI) Matlab to facilitate the calculation of prediction HDI values by each user. The research obtained the best prediction accuracy level using the learning rate (α) = 0.1, Target Error = 0.0000001, Maximum epoch = 500, network architecture 3-3-1, and scheme of data composition consists of 80% training data and 20% testing data. The percentage of prediction accuracy obtained for Maluku Province HDI is 99.57%, the average percentage of absolute error (MAPE) is 0.0043%. The pattern of predictive data shows an increase in HDI values from 2019-2023.