Open AccessMotivation of Students’ Persistency for Online Learning under Multiple Mediation Effect
Author(s) -
Yan Wang,
Peng Su,
Xiwen Liu,
Xin Zhao,
Fengming Jiao,
Guiling Liu,
Changtian Wang
Publication year - 2022
Publication title -
international journal of emerging technologies in learning/international journal: emerging technologies in learning
Language(s) - English
Resource type - Journals
eISSN - 1868-8799
pISSN - 1863-0383
DOI - 10.3991/ijet.v17i07.30399
Subject(s) - mediation , online learning , computer science , support vector machine , artificial intelligence , perspective (graphical) , machine learning , classifier (uml) , regression analysis , motivation to learn , psychology , mathematics education , multimedia , political science , law
This paper probed deep into the motivation of students’ persistency for online learning from the perspective of user experience of online learning platforms, in the purpose of increasing user stickiness and formulating effective operation strategies in a targeted manner. Existing studies on the motivation of students’ persistency for online learning mostly focus on theories, while few of them have talked about the problem with the multiple mediation effect taken into consideration, for this reason, this paper aims to fill in this research gap and explore the mechanism behind the motivation of students to carry out online learning persistently under the multiple mediation effect. At first, this paper built an improved support vector machine (SVM) classifier and used it to predict the duration of students' online learning; then, it adopted a structural equation model to analyze the data of students’ willingness to continue online learning; after that, this paper gave a theoretical analysis on the motivation of students’ persistency for online learning under multiple mediation effect, and constructed a basic regression model for the said matter; at last, this paper employed experimental results to verify the prediction accuracy of the constructed model, and gave the corresponding estimation results.