Integrasi Variable-Centered Intelligent Rule System dengan Teori Dempster-Shafer pada Sistem Pakar Infeksi Saluran Pernafasan Akut

Abstract

Acute Respiratory Infection (ARI) is a disease caused by infections of the respiratory tract, larynx, pharynx, sinuses and nose. ARI often causes death because the sufferer who comes for treatment is underestimated is already suffering from severe ARI. In 2013 to 2015 ARI was one of the ten most common illnesses in the city of Kupang, where ARI ranked first, followed by other diseases of the upper respiratory tract and grastitis. This study produced an expert system to diagnose ARI using the Variable Centered-Rule System method which functions to facilitate knowledge development and Dempster-Shafer Theory which serves to overcome uncertainty by entering the density of each symptom of ARI in the system. The VCIRS method is a method of building knowledge and inference strategies on expert systems. This method is rigid in accommodating changes in inference strategies except for changes in knowledge structures. This study aims to make the VCIRS method dynamic in an inference process where the sequence of variables in inference is determined by the occurrence and density of the variable. System accuracy by using medical record data of 95% with the triggering sequence of symptoms becoming dynamic every time a consultation session occurs.