Literaturnachweis - Detailanzeige
Autor/inn/en | Wang, Peipei; Li, Lin; Wang, Ru; Xie, Yifan; Zhang, Jianwei |
---|---|
Titel | Complexity-Based Attentive Interactive Student Performance Prediction for Personalized Course Study Planning |
Quelle | In: Education and Information Technologies, 27 (2022) 4, S.5471-5493 (23 Seiten)Infoseite zur Zeitschrift
PDF als Volltext |
Zusatzinformation | ORCID (Wang, Peipei) |
Sprache | englisch |
Dokumenttyp | gedruckt; online; Zeitschriftenaufsatz |
ISSN | 1360-2357 |
DOI | 10.1007/s10639-021-10842-3 |
Schlagwörter | Planning; Prediction; Academic Achievement; Grades (Scholastic); Academic Failure; Grade Point Average; Models; Tests; Difficulty Level; Attention |
Abstract | Planning course study is critical to facilitate strategic intervention in education. As a significant basis of planning course study, student performance prediction aims to utilize students existing relevant information to predict their future learning performance including course grades, course failure, grade point average, etc. We target course grade prediction for course study planning, provide reminders for students' learning and teachers' teaching. A student's performance is usually influenced by a variety of factors, such as the difficulty of the test, personal performance, and learning conditions. Recent studies focus on the various factors that influence student performance. However, most of them ignore the differences in the impact of these factors on different students. Moreover, existing prediction models pay less attention to the impact of test complexity on students' performance. To tackle these problems, we propose a novel Complexity-based Attentive Interactive Student Performance prediction model (CAISP) for personalized course study planning. In CAISP, the individual differences among students can be considered by assigning dynamically different importance using attention network to facilitate personalized planning. Especially, to effectively alleviate the student performance bias caused by test complexity, we fuse test complexity features to enhance the ability of feature representation via deep neural networks. Meanwhile, we combine factorization machines to develop a deep joint representation based on interaction learning and further boost the correlation between features, leading to optimal prediction results. Extensive experiments on two real-world datasets show that CAISP outperforms the state-of-the-art solutions. (As Provided). |
Anmerkungen | Springer. Available from: Springer Nature. One New York Plaza, Suite 4600, New York, NY 10004. Tel: 800-777-4643; Tel: 212-460-1500; Fax: 212-460-1700; e-mail: customerservice@springernature.com; Web site: https://link.springer.com/ |
Erfasst von | ERIC (Education Resources Information Center), Washington, DC |
Update | 2024/1/01 |