<![CDATA[Mental health problems in Indonesia are increasing, with university students being one of the groups vulnerable to depression due to academic pressure, social expectations, and exposure to negative information. Early detection of depression still relies on questionnaire methods that have limitations in objectivity and accuracy. Therefore, this research aims to develop a classification system for student depression using image recognition technology with Support Vector Machine (SVM). The system analyses students' facial expressions and combines them with questionnaire results to improve the accuracy of early depression detection. The results showed that out of 131 respondents, 74% experienced moderate depression, with academic pressure as the main factor. This finding is consistent with the condition of final-year students who face high academic loads. With this method, early detection of depression is more accurate than conventional methods, which can help intervene more quickly in dealing with student mental health crises.]]>
Copyrights © 2025
