Perangkat Lunak Untuk Memprediksi Penyakit Hati Jenis Hepatitis Menggunakan Algoritma Support Vector Machine (SVM)

Andres, Yohanes
Widhiarso, Wijang

MDP Student Conference
Vol 4 No 1 (2025): The 4th MDP Student Conference 2025

Hepatitis is an inflammation of the liver that can be caused by viral infections, genetic factors, alcohol, and medications. This disease is difficult to diagnose clinically due to its varying symptoms. This study aims to classify the types of hepatitis using the Support Vector Machine (SVM) algorithm. The method employed is SVM with various kernels, utilizing a dataset consisting of 620 records with 19 features and 1 class attribute. The data is split into 80% for training and 20% for testing. The results show that the polynomial kernel delivers the best performance with an accuracy of 96%, precision of 100%, recall of 81%, and F1-score of 89%. The conclusion of this study is that the SVM method with a polynomial kernel can serve as an effective tool in detecting the presence or absence of hepatitis.