Early diagnosis of skin diseases which is caused by fungal infections is necessary to reduce the risk of disease. In this article, an expert system that implements the Bayes’ theorem is proposed as a solution for early diagnosis of skin diseases, especially those caused by fungal infections. Knowledge acquisition from experts, in this case a specialist in skin and venereal diseases, is carried out to obtain a knowledge base. A total of 31 data from medical records were used in this study. Types of skin diseases that can be detected are tinea korporis, tinea unguium, pitiriasis versicolor, and mild mucocutaneous candidosis. Bayes’ theorem is implemented in the inference engine, resulting in a probability of the type of skin disease suffered based on symptom inputs from the user. Results, which are produced by the system, are then validated by expert. Based on the results of the validation, 27 results matched the experts’ validation while 4 results did not, so that the level of system accuracy based on the validation results was 87.1%.
Copyrights © 2021