IMPLEMENTASI METODE PEMBOBOTAN BERBASIS ATURAN DAN METODE PROFILE MATCHING PADA SISTEM PAKAR MEDIS UNTUK PREDIKSI RISIKO HIPERTENSI

Abstract

Cardiovascular is a disease that often causes death. One of the cardiovascular diseases that often cause death is the risk of Hipertensi. The highest risk factors for premature death and disability in the world are caused by smoking habits, high systolic blood pressure, and increased blood sugar levels. This death factor is because people with Hipertensi generally do not experience any symptoms until their blood pressure is too high which can cause death. Efforts that can be made are by utilizing information technology in the form of a medical expert system to Kelasify the risk of Hipertensi. This study aims to develop a medical expert system in a different way using rule-based weighting methods and profile matching. The weighting method is used to determine the risk weight based on patient variables, while the profile matching method is used to calculate the risk Kelasification based on the core factor and secondary factor variables on the risk of Hipertensi. System evaluation is carried out by comparing asset data taken from the Pima Indian Hipertensi Data (NHANES) with results from the system. The results of the comparison show that the accuracy of the proposed system is 96.67%. The proposed system is also compared with other Kelasification methods such as decision tree, Random Tree, Decision Stump, KNN, Naïve BaYa, Deep Learning, and Rule Induction. Based on the comparison results, the proposed system has a better level of accuracy, therefore the system developed can be used to Kelasify risks for other types of diseases.