Prediksi Kelayakan Sumber Air Minum Menggunakan Algoritma Support Vector Machine (SVM)

Abstract

Water is an essential need for living things and also a source of energy. Where almost all living things in this world really need water because there is not a single creature in this world that does not contain water. With this in mind, it is necessary to predict the feasibility of water sources to find out whether the water is suitable for consumption or not. Therefore it is necessary to carry out this research with the aim of obtaining water quality from water sources that are suitable for consumption. In predicting the feasibility of this water source, the Support Vector Machine (SVM) classification method will be used, which is a machine learning algorithm that is used for classification or regression problems that have been widely used with effective results, good accuracy, powerful and flexible, and can be used in many applications. . To obtain accuracy based on data on the quality of water sources, it can be categorized based on water PH, conductivity, organic carbon and other contents. This research produces an accuracy of the feasibility of water sources where the accuracy results are 71% for SVM, 61% for decision trees, 67% for random forest based on the accuracy above can help analyze and classify water feasibility in Indonesia.