Klasifikasi Penyakit Menular Dengan Algoritma Machine Learning Berbasis SVM

Abstract

Infectious diseases pose a serious threat to public health, especially with their rapid spread and the difficulty of detecting early symptoms in some cases. Accurate classification of infectious diseases is essential to support early diagnosis and appropriate treatment. In this research, a machine learning algorithm based on Support Vector Machine (SVM) was used to classify types of infectious diseases. This method was chosen because of its ability to handle complex datasets and produce good classification, especially on data with non-linear patterns. This research uses infectious disease datasets from trusted sources which are processed using the Knowledge Discovery in Databases (KDD) method for extracting relevant features. Several SVM kernels, namely linear, radial basis function (RBF), sigmoid, and polynomial, were evaluated to determine the most optimal kernel in increasing classification accuracy. The aim of this research is to identify the most effective method in predicting infectious diseases, so that it can be applied in decision support systems in the health sector. The research results show that the polynomial kernel provides the highest accuracy compared to other kernels, with an accuracy level reaching 75%. With these results, it is hoped that the SVM-based classification model ca be a solution in identifying and treating infectious diseases more efficiently.