The comparison of accuracy on classification data with machine learning algorithms (Case study: Human development index by regency/city in Indonesia 2020)

Machine learning algorithms have become more popular in a multi sectors due to its high accuracy. The aim of this study was to compare the accuracy on classification data with machine learning algorithms. The Human Development Index indicator, by regencies/cities in Indonesia in 2020 and its constituent components, namely Life Expectancy at Birth, Expected Years of Schooling, Mean Years of Schooling, and Expenditures Per Capita, were utilized as the case study. Machine learning algorithms applied are Linear Discriminant Analysis, Classification and Regression Trees, k-Nearest Neighbors, Support Vector Machines with radial basis function kernel, and Random Forest. Each model use 10-fold cross validation and executed ten times in total to estimate accuracy. The Support Vector Machine, which has an accuracy of 0.959 on the training data and 0.9604 on prediction of testing data, has the best accuracy based on the average accuracy