Analisis Komparatif Particle Swarm Optimization (Pso) Dan Genetika Algoritma (Ga) Untuk Meningkatkan Algoritma Naïve Bayes Dalam Memprediksi Penyakit Hepatitis

Abstract

The aims of this  study is to predict the types of hepatitis, by predicting what types of hepatitis suffered. The results of the study used the Partile Swarm Optimization (PSO) based naïve bayes algorithm and the naïve bayes algorithm based on the Genetic Algorithm (GA). The results of the accuracy of naïve Bayes without optimization are 83.08% and the AUC value is 0.826, using PSO optimization of 84.50% and AUC value of 0.883, while naïve Bayes uses GA optimization of 85.79% and an AUC value of 0.901. From this study that naïve Bayes using GA optimization got the highest accuracy value with an increase of 2.71% in predicting hepatitis.