Comparison of k-Nearest Neighbor (k-NN) and Support Vector Machine (SVM) Methods for Classification of Poverty Data in Papua

Abstract

Classification is a job of assessing data objects to include them in a particular class from a number of available classes. The classification method used is the k-Nearest Neighbor (k-NN) and Support Vector Machine (SVM) methods. The data used in this study is data on poverty in Papua with the category of the number of low/high level poor people. Of the 29 regencies/cities that were sampled, 15 regencies/cities represent the number of low-level poor people and 14 districts/cities are the number of high-level poor people. The results of the analysis obtained are the k-Nearest Neighbor (k-NN) method with a value of k=15 producing an accuracy of 58.62%, while the Support Vector Machine (SVM) method with Parameter cost = 1 using the RBF kernel produces an accuracy value. by 93.1%. The classification criteria to find the best method is to look at the Root Mean Square Error (RMSE) which states that the Support Vector Machine (SVM) method is better than the k-Nearest Neighbor (k-NN) method.