<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>In this paper, a mental health evaluation system based on KB-SMOTE algorithm and XGBoost model was successfully established. Firstly, the questionnaire is reasonably set up and distributed to the college students majoring in physical education to fill in the questionnaire, to obtain the college students’ mental health questionnaire survey data. Secondly, the KB-SMOTE algorithm is used to optimize the data of questionnaire tuning data. The training set was utilized to establish the primary parameters of the XGBoost model. The final parameters were determined by the parameters that correspond to the highest accuracy of the XGBoost model. Finally, the model is built based on the best parameters obtained from the training set, and the optimized test data set is brought into the model to output the value of college students’ mental health assessment. The results show that the model performs best when the learning rate is 0.02, the tree depth is 5, the maximum number of iterations is 500, and the L2 regular term is 5. The XGBoost model predicts 4160 cases correctly and 286 cases incorrectly, with an overall correctness rate of 93.6%, and its corresponding ROC curve is closer to the y-axis, with an AUC value of 0.9154. By utilizing the XGBoost model, the model can accurately determine the mental health of college students. The mental health of college students can be accurately evaluated by models.</p>
</abstract>

In this paper, a mental health evaluation system based on KB-SMOTE algorithm and XGBoost model was successfully established. Firstly, the questionnaire is reasonably set up and distributed to the college students majoring in physical education to fill in the questionnaire, to obtain the college students’ mental health questionnaire survey data. Secondly, the KB-SMOTE algorithm is used to optimize the data of questionnaire tuning data. The training set was utilized to establish the primary parameters of the XGBoost model. The final parameters were determined by the parameters that correspond to the highest accuracy of the XGBoost model. Finally, the model is built based on the best parameters obtained from the training set, and the optimized test data set is brought into the model to output the value of college students’ mental health assessment. The results show that the model performs best when the learning rate is 0.02, the tree depth is 5, the maximum number of iterations is 500, and the L2 regular term is 5. The XGBoost model predicts 4160 cases correctly and 286 cases incorrectly, with an overall correctness rate of 93.6%, and its corresponding ROC curve is closer to the y-axis, with an AUC value of 0.9154. By utilizing the XGBoost model, the model can accurately determine the mental health of college students. The mental health of college students can be accurately evaluated by models.

Construction of Mental Health Evaluation System for College Students’ Physical Education Learning in the Internet Era

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

In this paper, a mental health evaluation system based on KB-SMOTE algorithm and XGBoost model was successfully established. Firstly, the questionnaire is...

Construction of Mental Health Evaluation System for College Students’ Physical Education Learning in the Internet Era

Online veröffentlicht: 31. Jan. 2024

Seitenbereich: -

Eingereicht: 22. Dez. 2023

Akzeptiert: 01. Jan. 2024

DOI: https://doi.org/10.2478/amns-2024-0246

Schlüsselwörter
KB-SMOTE algorithm, XGBoost model, Mental health, Evaluation system, Physical education college

© 2024 Yu Hong, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Construction of Mental Health Evaluation System for College Students’ Physical Education Learning in the Internet Era

Online veröffentlicht: 31. Jan. 2024

Seitenbereich: -

Eingereicht: 22. Dez. 2023

Akzeptiert: 01. Jan. 2024

DOI: https://doi.org/10.2478/amns-2024-0246

SchlüsselwörterKB-SMOTE algorithm, XGBoost model, Mental health, Evaluation system, Physical education college

© 2024 Yu Hong, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Schlüsselwörter
KB-SMOTE algorithm, XGBoost model, Mental health, Evaluation system, Physical education college