Depression anxiety stress scale and handgrip using machine learning analysis

• Main Authors: Usman, Sahnius, Rusli, Fatin ‘Aliah, A. Jalil, Siti Zura, Bani, Nurul Aini
• Format: Conference or Workshop Item
• Published: 2022
• Subjects: T Technology (General)
• Online Access: http://eprints.utm.my/id/eprint/98915/; http://dx.doi.org/10.1109/ICSSA54161.2022.9870948

Stress is an emotional or physical state of tension. Stress is the body's natural response to difficulty or a great deal of work. Each of us has a unique reaction to stress. Our capacity for adaptation can be influenced by our genetics, early life events, personality, and socioeconomic situations. This study used handgrip strength (HGS) reading for stress level screening together with Depression Anxiety Stress Scale (DASS) as an early assessment tool. This data of DASS and HGS were analyzed using Random Forest and Support Vector Machine. The dataset is normalized between 0 to 1 due to different units in different measurement tools. The result shows that Random Forest gives an accuracy of 93.75%, a specificity of 94.90%, and a sensitivity of 93.80%. However, SVM gives 87.50% accuracy, 90.30% specificity, and 87.50% sensitivity. This concludes that the Random Forest is better than SVM in terms of stress level classification.