Research on the algorithm-driven adaptive education model and its impact on students' learning motivation and engagement
Data and Metadata · 15.05.2026 · 5 · 782
DOI: 10.56294/dm2026782 ↗Kurz gefasst
Fachartikel in der Zeitschrift Data and Metadata, online erschienen am 15.05.2026. Verfasst von Su Peng und 1 weiteren Autor:innen. Die Arbeit nennt 18 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Introduction
Algorithm-based intelligence is a rapidly emerging field in modern educational systems that seeks to boost personalized learning and student motivation and engagement within a digital environment. Current methods are focused on predictive modelling and learner static profiling, with less focus on adaptability and tracking of learner interactions; this is a missed opportunity for context-aware adaptive learning systems to support improved learner interaction and outcomes.
Objective
The goal of this research is to create an algorithm-empowered adaptive education model which can boost student motivation and engagement by using intelligent prediction and adaptive feedback mechanisms and deep sequence learning architectures.
Method
The adaptive education engagement dataset consists of 9,798 records and 30 features with engagement and behavioral attributes, and Min-Max normalization is used for feature scaling, and Principal Component Analysis (PCA) is used for dimensionality reduction to further improve the computational efficiency and reduce the redundancy. A model called Sunflower Optimization-driven Adaptive Long Short-Term Memory (SO-ALSTM) is proposed to model the sequential dependencies of the learner's engagement and motivational patterns, and hyperparameters are optimized to ensure the stability of convergence while improving the model's performance from behavioral features for accurately predicting learning states. The SO-ALSTM model provides strong and effective modeling about adaptive learning behavior in complex and irregular educational interaction situation.
Results
The experiments were conducted in Python and the results show that the proposed SO-ALSTM model outperforms the existing models by achieving 98,66 % accuracy, 98,68 % precision, 98,14 % recall and 98,41 %[w2.1][w2.2] F1-score.
Conclusions
The results have demonstrated the effectiveness of the proposed approach to improving the accuracy of learners' engagement prediction and analysis of motivation, which contributes to the development of adaptive and personalized learning systems in a digital learning environment.
Zitieren
APA
Peng, S., & Li, X. (2026). Research on the algorithm-driven adaptive education model and its impact on students' learning motivation and engagement. Data and Metadata, 5, 782. https://doi.org/10.56294/dm2026782
Vancouver
Peng S, Li X. Research on the algorithm-driven adaptive education model and its impact on students' learning motivation and engagement. Data and Metadata. 2026;5:782. doi:10.56294/dm2026782
BibTeX
@article{Peng2026,
author = {Peng, Su and Li, Xiaolong},
title = {Research on the algorithm-driven adaptive education model and its impact on students' learning motivation and engagement},
journal = {Data and Metadata},
year = {2026},
volume = {5},
pages = {782},
doi = {10.56294/dm2026782}
}
EUVIMED