Sistem Pakar Diagnosa Penyakit Kulit Menggunakan Metode Forward Chaining dan Certainty Factor

Abstract

Skin diseases have a variety of symptoms and often show similarities, making it difficult for the public to recognize the type of disease and determine the appropriate initial steps. This research aims to design and build an expert system for diagnosing skin diseases in humans at the Restu Ibu Clinic by combining the Forward Chaining and Certainty Factor methods. Forward chaining is used to trace facts in the form of symptoms and match them with IF-THEN rules in the knowledge base, while the certainty factor is used to calculate the confidence level of the diagnosis based on the user's confidence value and expert weight. Data collection was carried out through observation, interviews with medical staff, and literature study. The knowledge base consists of 10 skin diseases, 42 symptoms, and 17 rules. The system is developed using PHP and MySQL and provides features for managing diseases, symptoms, rules, accounts, consultations, consultation history, and result printing. Testing on five case studies produced confidence values of 97.05% for psoriasis, 99.55% for atopic dermatitis, 98.53% for urticaria, 99.11% for tinea, and 100% for scabies. The values represent the confidence level of the inference results, not a measure of clinical accuracy. Black Box testing on 59 scenarios showed that all functions operated as expected. The system can be used as an initial screening aid and information support before further examination by medical personnel.