| Abstract |
| This study aimed to predict psychopathology based on personality measures via supervised machine learning methodology. We implemented the Singer-Loomis Type Deployment Inventory (SLTDI) for psychological typology and the Korean version of the Revised Symptom Checklist 90 (KSCL-95) for psychopathology. A total of 521 Korean adults from across the country participated in the online survey. Statistical analyses including correlation, k-means cluster analysis, classification, and regression-based decoding were performed. Results revealed four differentiated clusters on the spectrum of clinical severity. Moreover, SLTDI could distinguish between hypothesis-driven and data-driven clusters by chance. KSCL-95’s three subcategories, as well as its validity, were accurately classified. Regression-based decoding results showed that their typology data significantly predicted social desirability, depression, anxiety, obsessive-compulsive disorder, PTSD, schizophrenia, stress vulnerability, and interpersonal sensitivity significantly. Overall, these findings suggest that personality tests could be utilized to screen for the severity of psychopathology and to implement prevention and early intervention strategies. |
|
|
| Key Words |
| Psychological Typology, Machine Learning, Singer-Loomis Type Deployment Inventory, Personality, Mental Health, KSCL-95, 심리 유형, 기계학습, 싱어루미스 심리 유형 검사, 성격, 정신병리, 간이정신진단검사 |
|
|
 |
|
