<![CDATA[Academic stress is a common problem experienced by final semester students during the thesis preparation process. However, many students are unaware of their stress levels due to limited access to campus counseling services and social stigma regarding mental health issues. This study aims to develop a web-based expert system for diagnosing stress levels in final semester students using the Certainty Factor method. The research methodology adopts the Turban model consisting of four stages: intelligence, design, choice, and implementation. Data were collected through an online questionnaire distributed to 60 final semester student respondents from various universities in Indonesia. The system was developed using the PHP programming language with a MySQL database and implements the Certainty Factor (CF) method for diagnosis calculation. The CF method works by combining expert CF values and user CF values iteratively until a final CF value is obtained, which is classified into three stress level categories: mild, moderate, and severe. The results showed that the system successfully diagnosed 54 students (90.0%) with severe stress, 2 students (3.3%) with moderate stress, and 4 students (6.7%) with mild stress. Black Box Testing on 16 scenarios showed that all system functions were valid (100%). Comparison of system diagnoses with psychologist expert diagnoses on 60 respondents showed an accuracy rate of 100%. This system can be used as an early detection tool for academic stress independently while maintaining user privacy.]]>
Copyrights © 2026
