Article: Mental toughness prediction model of college students based on optimal elastic network regression Journal: International Journal of Information and Communication Technology (IJICT) 2024 Vol.25 No.10 pp.19 - 33 Abstract: Predicting the level of mental toughness can help colleges and universities better understand the psychological condition of college students. This paper designs a prediction model of college students' mental toughness based on optimised elastic network regression (ENR) to address the redundant features as well as the overfitting problems of existing studies. Firstly, the ENR is optimised using Bayesian optimisation algorithm (BOENR). Secondly, the important influencing factors are extracted to the maximum extent by using the partial least squares method. Then, linear discriminant analysis (LDA) is used for feature screening of key influencing factors, Pearson's correlation coefficient is used to measure the redundancy relationship among features, and finally, BOENR estimation of regression coefficients is computed based on each feature sample separately. The experimental outcome indicates that the MSE and MAE of the designed model are reduced by 0.0395-0.2264 compared with the other five models. Inderscience Publishers - linking academia, business and industry through research

Title: Mental toughness prediction model of college students based on optimal elastic network regression

Authors: Fulian Liu

Addresses: Mental Health Education Center for College Students, Wuxi Institute of Technology, Wuxi, 214121, China

Abstract: Predicting the level of mental toughness can help colleges and universities better understand the psychological condition of college students. This paper designs a prediction model of college students' mental toughness based on optimised elastic network regression (ENR) to address the redundant features as well as the overfitting problems of existing studies. Firstly, the ENR is optimised using Bayesian optimisation algorithm (BOENR). Secondly, the important influencing factors are extracted to the maximum extent by using the partial least squares method. Then, linear discriminant analysis (LDA) is used for feature screening of key influencing factors, Pearson's correlation coefficient is used to measure the redundancy relationship among features, and finally, BOENR estimation of regression coefficients is computed based on each feature sample separately. The experimental outcome indicates that the MSE and MAE of the designed model are reduced by 0.0395-0.2264 compared with the other five models.

Keywords: mental toughness prediction; elastic network regression; ENR; Bayesian optimisation; partial least square; linear discriminant analysis; LDA.

DOI: 10.1504/IJICT.2024.143331

International Journal of Information and Communication Technology, 2024 Vol.25 No.10, pp.19 - 33

Received: 12 Oct 2024
Accepted: 28 Oct 2024
Published online: 13 Dec 2024 *

Title: Mental toughness prediction model of college students based on optimal elastic network regression

Keep up-to-date