<![CDATA[Mental health is an important aspect that affects the ability of final-semester students to complete their studies and face academic and non-academic pressures. The problem that arises is the difficulty in assessing mental health conditions because it is subjective and complex. The Tsukamoto fuzzy method is used because it is able to handle data uncertainty and provides results in the form of measurable crisp values. This study aims to apply the Tsukamoto fuzzy logic method in determining the level of mental health of final-semester students in a more objective and measurable manner. This system model uses four input variables, namely stress level, sleep quality, emotional exhaustion, and duration of gadget use, with 81 rules (rule base) that form relationships between variables. The inference process is carried out through the stages of fuzzification, rule inference, and defuzzification with increasing and decreasing linear triangular membership functions. Testing was carried out using MATLAB by comparing the prediction results to actual data to calculate the model accuracy level using the Mean Absolute Percentage Error (MAPE). The results showed that the total MAPE value was 19.34%, which is in the range of 10%–20% so it is included in the good accuracy category. This demonstrates that the Tsukamoto fuzzy method can provide fairly accurate predictions of the mental health of final-semester students. Therefore, this system can be used as a tool for evaluating and early detection of student mental health in higher education settings.]]>
Copyrights © 2025
