Comparison of Case-Based Reasoning and Certainty Factor Methods for Dengue Diagnosis

Abstract

Dengue Hemorrhagic Fever (DHF) is a disease caused by a dengue virus infection that is transmitted through the bite of Aedes Aegypti and Aedes Albopictus mosquitoes. People with dengue fever who are diagnosed based on symptoms that appear without any medical considerations can lead to mishandling. Therefore, a computer-based system is needed to overcome these problems. This study aims to compare two methods: Case-Based Reasoning (CBR) and Certainty Factor (CF) for diagnosing dengue. The CBR method computed the similarity value of the new case disease with the previous case to justify the disease. Meanwhile, the CF method used the certainty value of the disease. The dataset consists of 110 patient data divided into three categories of dengue fever, namely, dengue fever, dengue hemorrhagic fever, and dengue shock syndrome. There are 19 symptoms of the patient recorded. The expert system presents the result of DHF diagnosis, the appropriate treatment solutions, and the comparison results of the diagnosis from the CBR and CF methods. Based on the evaluation result, it shows that the CBR method achieves the value of accuracy, precision, and recall is 84.455%, 84.455%, and 100%, while the CF method obtain 80.909%, 80.909%, and 100%, respectively. Based on the test results, it shows that there are differences in values, so that the conclusion is that the CBR method is more accurate in diagnosing DHF.