<![CDATA[Mental health issues such as stress, anxiety, and depression are increasingly affecting individuals across various levels, from mild to severe. Despite this, public perception often underestimates these conditions, particularly depression, which is frequently misunderstood as a temporary emotional phase. In fact, depression is a serious mental disorder that requires proper attention and treatment. This study aims to classify early patterns of depressive symptoms using the Agglomerative Hierarchical Clustering (AHC) algorithm. Data were collected from 1,030 respondents using the PHQ-9 questionnaire, a widely recognized tool for assessing depression levels. The analysis was conducted using the AHC method, and cluster evaluation was performed using the Davies-Bouldin Index (DBI) to determine optimal clustering. Results indicate that the Complete Linkage method provided the most optimal performance, as reflected in the highest Cophenetic Correlation Coefficient compared to other linkage methods. The DBI evaluation suggests that the optimal number of clusters is four. Among the respondents, 17.2% fall into cluster 1, 32.5% into cluster 2, 21.9% into cluster 3, and 28.3% into cluster 4. These clusters represent distinct patterns of depressive symptoms, which can be used for early detection and improved mental health interventions.]]>
Copyrights © 2025
