Khairunnisa, Khairunnisa (2026) A COMPARISON OF THE PERFOMANCE OF THE XGBOOST AND CATBOOST ALGORITHMS IN CLASSIFYING THE LEVELS OF DEPRESSION AMONG COLLEGE STUDENTS. Undergraduate thesis, UPN VETERAN JAWA TIMUR.
![[img]](https://repository.upnjatim.ac.id/style/images/fileicons/text.png) |
Text (Cover)
22081010325_Cover.pdf Download (991kB) |
|
Text (Bab I)
22081010325_Bab I.pdf Download (254kB) |
|
Text (Bab II)
22081010325_Bab II.pdf Restricted to Repository staff only until 7 September 2028. Download (787kB) | Request a copy |
|
Text (Bab III)
22081010325_Bab III.pdf Restricted to Repository staff only until 7 September 2028. Download (720kB) | Request a copy |
|
Text (Bab IV)
22081010325_Bab IV.pdf Restricted to Repository staff only until 7 September 2028. Download (854kB) | Request a copy |
|
Text (Bab V)
Bab V.pdf Download (215kB) |
|
Text (Daftar Pustaka)
2208101035_Daftar Pustaka.pdf Download (253kB) |
|
Text (Lampiran)
Lampiran.pdf Restricted to Repository staff only until 7 September 2028. Download (903kB) | Request a copy |
Abstract
Students' psychological health and academic performance may be impacted by depression, a mental health condition. This study compares the effectiveness of the XGBoost and CatBoost algorithms and attempts to categorize the depression levels of students. 306 senior undergraduate students made up the data used, and the PHQ-9, GAD-7, and ERQ instruments provided the characteristics. Four categories of depression were identified: normal, mild, moderate, and severe depression. Three data splitting scenarios (80:20, 70:30, and 60:40) were used in the investigation, coupled with SMOTE-NC and GridSearchCV hyperparameter tuning. With accuracy of 0.89, precision of 0.91, recall of 0.81, F1-score of 0.84, and AUC of 0.97, the findings demonstrated that XGBoost performed best at the 80:20 ratio. CatBoost, on the other hand, had an accuracy of 0.85 and an AUC of 0.98 at the 60:40 ratio. The best model for categorizing students' depression levels, according to the total evaluation results, was XGBoost with an 80:20 ratio
| Item Type: | Thesis (Undergraduate) | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Contributors: |
|
||||||||||||
| Subjects: | B Philosophy. Psychology. Religion > BF Psychology T Technology > TK Electrical engineering. Electronics Nuclear engineering > TK5105.888 World Wild Web |
||||||||||||
| Divisions: | Faculty of Computer Science > Departemen of Informatics | ||||||||||||
| Depositing User: | Khairun nisa | ||||||||||||
| Date Deposited: | 07 Sep 2026 02:15 | ||||||||||||
| Last Modified: | 07 Sep 2026 02:54 | ||||||||||||
| URI: | https://repository.upnjatim.ac.id/id/eprint/48507 |
Actions (login required)
 |
View Item |