<![CDATA[Mental disorders are health problems that require early detection and appropriate treatment. This study developed a classification model for mental disorders using the TabularBERT method and evaluated the effect of Principal Component Analysis (PCA) in improving performance. A mental disorder symptom dataset was used with a train-test split scheme, and evaluation was conducted using accuracy, precision, recall, F1-score metrics, and confusion matrix visualization. Model interpretability was analyzed using Local Interpretable Model-Agnostic Explanations (LIME). The results show that the application of PCA consistently improves model performance. LIME analysis revealed differences in feature contributions between models with and without PCA. This study confirms that the combination of TabularBERT, PCA, and LIME not only produces a high-performance classification model but also supports interpretability, making it potentially applicable in decision support systems in the field of mental health to improve the quality of mental disorder detection and treatment.]]>
Copyrights © 2025
