KOMPARASI ALGORITMA BERBASIS NEURAL NETWORK DALAM MENDETEKSI PENYAKIT HEPATITIS

Abstract

Hepatitisis aninfectious disease of the liverand is very dangerous in the world, several studies have been conducted to correctly diagnose patients is unknown but the most accurate method in predicting disease hepatitis in patients. In this study a comparison algorithm Multilayer Perceptron (MLP) and Support Vector Machine (SVM) algorithms to determine the most accurate in predicting disease hepatitis. This study uses secondary data in the form of hepatitis disease data obtained from the University of California Irvine Machine Learning repository of data. Testing the algorithms are carried out by using software that is known that rapidminer algorithm Support Vector Machine (SVM) has the highest value of accuracy is 90.64%, while the algorithm Multilayer Perceptron (MLP) has an accuracy of 84.38%. Thus the algorithm Support Vector Machine (SVM) to predic the hepatitis disease better.