Implementation of the Support Vector Machine (SVM) Method for Classification of Underdeveloped Areas in Indonesia

Development disparity is one of the common issues in developing countries, including Indonesia. According to Presidential Regulation of Indonesia Number 63 year 2020, several regions classified as underdeveloped from 2020 to 2024 are located in Sumatra, Nusa Tenggara, Sulawesi, Maluku, and Papua. Determining underdeveloped areas can be assessed through underdeveloped area indicators. The objective of this research is to ascertain the classification of underdeveloped areas in Indonesia based on seven underdeveloped area indicators in 2022. One of the methods that can be used for classification is Support Vector Machine (SVM). SVM is a supervised machine learning method typically used for classification. The reason for using the SVM method in this research is because SVM has a more mature concept compared to other classification algorithms and can handle both linear and non-linear classification problems. By using the SVM method, it was found that the best kernel function is the RBF kernel function, which can classify with an accuracy rate of 94.17%. There are 6 regencies initially categorized as not underdeveloped becoming underdeveloped based on the result of SVM, namely Situbondo Regency, Bangkalan Regency, Southwest Sumba Regency, Majene Regency, Merauke Regency, and Jayawijaya Regency. Meanwhile, there are no regencies initially categorized as underdeveloped becoming not underdeveloped.