PENERAPAN JARINGAN SYARAF TIRUAN UNTUK INDEKS PEMBANGUNAN MANUSIA MENGGUNAKAN ALGORITMA PERCEPTRON

Abstract

This study aims to analyze and predict the level of the Human Development Index (HDI) using the Artificial Neural Network (ANN) method with the Perceptron algorithm. The problem underlying this research is the large amount of data and the numerous variables affecting HDI, making manual analysis less effective and time-consuming. The data used in this study include Life Expectancy at Birth, Expected Years of Schooling, Mean Years of Schooling, and Adjusted Expenditure per Capita. The research was conducted using Human Development Index data obtained from the Family Information System (SIGAThe dataset was divided into training data and testing data to evaluate the model's ability to predict HDI levels into two categories, namely high and low. The system was developed using the PHP programming language and MySQL database and was designed using Unified Modeling Language (UML).The results of this study indicate that the Artificial Neural Network method with the Perceptron algorithm is capable of predicting Human Development Index levels effectively based on the available data. The Perceptron model was able to recognize the relationship patterns among Life Expectancy at Birth, Expected Years of Schooling, Mean Years of Schooling, and Adjusted Expenditure per Capita variables with HDI levels. The testing results produced a final Mean Squared Error (MSE) value of 0.141227 with an accuracy rate of 80.00% after the training process stopped at the 115th epoch with a learning rate of 0.1. The developed system can assist in the analysis and prediction of human development levels and can be used as a decision-support tool in human development planning.