Enhancing smart healthcare with female students’ stress and anxiety detection using machine learning DOI
Farhad Hosseinzadeh Lotfı, Ahmad Lotfi,

Masoud Lotfi

и другие.

Psychology Health & Medicine, Год журнала: 2025, Номер unknown, С. 1 - 20

Опубликована: Март 30, 2025

Machine learning (ML) is widely used to predict and detect stress anxiety. Early detection of or anxiety crucial for clinical pathways enhance the supportive environment in society, particularly among female students. This study aims assess improve accuracy detecting students using machine algorithms functions. Three primary features are cigarette smoking, physical activity grade point average (GPA). The multiple linear regression analysis conducted on 160 datasets obtained from State-Trait Anxiety Inventory (STAI) at University Belgrade was selected. A heat map utilised identify least engaging areas model along with most state factors. Additionally, R-squared (R2), mean absolute error (MAE), squared (MSE) root (RMSE) were employed errors both pre-intervention post-intervention, focusing key related students' Using K-Means algorithm, cluster executed samples (N = 160) three features. total score 44.39% (out 80%) considered moderate. indicated a strong relationship between variables. Overall, post-intervention stage yielded acceptable results compared stage. Two clusters identified, demonstrating that these can accurately research analyse better algorithm. ML functions demonstrated smoking cigarettes, GPA have reduced during detection.

Язык: Английский

Enhancing smart healthcare with female students’ stress and anxiety detection using machine learning DOI
Farhad Hosseinzadeh Lotfı, Ahmad Lotfi,

Masoud Lotfi

и другие.

Psychology Health & Medicine, Год журнала: 2025, Номер unknown, С. 1 - 20

Опубликована: Март 30, 2025

Machine learning (ML) is widely used to predict and detect stress anxiety. Early detection of or anxiety crucial for clinical pathways enhance the supportive environment in society, particularly among female students. This study aims assess improve accuracy detecting students using machine algorithms functions. Three primary features are cigarette smoking, physical activity grade point average (GPA). The multiple linear regression analysis conducted on 160 datasets obtained from State-Trait Anxiety Inventory (STAI) at University Belgrade was selected. A heat map utilised identify least engaging areas model along with most state factors. Additionally, R-squared (R2), mean absolute error (MAE), squared (MSE) root (RMSE) were employed errors both pre-intervention post-intervention, focusing key related students' Using K-Means algorithm, cluster executed samples (N = 160) three features. total score 44.39% (out 80%) considered moderate. indicated a strong relationship between variables. Overall, post-intervention stage yielded acceptable results compared stage. Two clusters identified, demonstrating that these can accurately research analyse better algorithm. ML functions demonstrated smoking cigarettes, GPA have reduced during detection.

Язык: Английский

Процитировано

0