We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Machine learning analysis of factors affecting college students' academic performance.
- Authors
Lu, Jingzhao; Liu, Yaju; Liu, Shuo; Yan, Zhuo; Zhao, Xiaoyu; Zhang, Yi; Yang, Chongran; Zhang, Haoxin; Su, Wei; Zhao, Peihong
- Abstract
This study aims to explore various key factors influencing the academic performance of college students, including metacognitive awareness, learning motivation, participation in learning, environmental factors, time management, and mental health. By employing the chi-square test to identify features closely related to academic performance, this paper discussed the main influencing factors and utilized machine learning models (such as LOG, SVC, RFC, XGBoost) for prediction. Experimental results indicate that the XGBoost model performs the best in terms of recall and accuracy, providing a robust prediction for academic performance. Empirical analysis reveals that metacognitive awareness, learning motivation, and participation in learning are crucial factors influencing academic performance. Additionally, time management, environmental factors, and mental health are confirmed to have a significant impact on students' academic achievements. Furthermore, the positive influence of professional training on academic performance is validated, contributing to the integration of theoretical knowledge and practical application, enhancing students' overall comprehensive competence. The conclusions offer guidance for future educational management and guidance, emphasizing the importance of cultivating students' learning motivation, improving participation in learning, and addressing time management and mental health issues, as well as recognizing the positive role of professional training.
- Subjects
MACHINE learning; EDUCATIONAL counseling; ACADEMIC achievement; TIME management; FACTOR analysis
- Publication
Frontiers in Psychology, 2025, p1
- ISSN
1664-1078
- Publication type
Academic Journal
- DOI
10.3389/fpsyg.2024.1447825