Analysis of dengue fever diagnosis based on patient symptoms using the certainty factor method

Abstract

Dengue fever is a major public health problem, especially in tropical countries such as Indonesia. The similarity of symptoms with other diseases often makes early diagnosis difficult, leading to delays in proper treatment. To address this issue, an expert system can be utilized to assist in early diagnosis based on patient symptoms. An expert system is a branch of artificial intelligence that adopts expert knowledge to solve specific problems. One of the methods used to handle uncertainty in diagnosis is the Certainty Factor (CF) method, which represents the level of confidence of an expert in a given hypothesis. This study aims to analyze the diagnosis of dengue fever based on patient symptoms using the Certainty Factor method. The data used in this study were obtained from a Kaggle dataset containing symptoms such as fever, headache, joint pain, and bleeding. These symptoms are used to form a knowledge base and diagnostic rules in the expert system. The CF method is applied to calculate the confidence level of diagnosis based on the combination of expert knowledge and user input. The results show that the proposed system is able to provide a diagnosis with a measurable confidence value, thereby supporting early detection of dengue fever. This research is expected to contribute to the development of intelligent decision support systems in the medical field.