.png){kind=icon}
.png){kind=icon}
Return
Optimizing personalised recommender systems for teachers’ digital learning models using deep learning algorithms
.png){kind=icon}
PRE
AI
DOI：10.1080/10494820.2025.2528104.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Digital Learning Models for Teachers is an important educational development to improve teaching effectiveness through digital technology. It creates learning experiences with personalization and customization by providing teachers with more accurate feedback on their teaching. However, the current problem is that the learning question types must be recommended according to the students’ needs, leading to inadequate learning effects. To solve the problem, this paper proposed a personalized recommendation algorithm based on a graph neural network for teachers’ digital learning models (PRAGNN). In particular, firstly, integrating DINA cognitive diagnosis and gray partial correlation evaluation is used to construct a student model by modeling students’ mastery of knowledge points and cognitive ability level. Secondly, a graph convolutional neural network is introduced and combined with the sequential relationship between subject knowledge points to automatically capture the semantic information of higher-order structures in the knowledge points to achieve personalized recommendations. Ultimately, the algorithm achieved 85.60% accuracy through comparative experiments. This research provides a new idea for constructing a personalized recommendation system for the digital learning model of teachers, it has significant value for the sustainable development of lifelong learning that promotes mutual growth between teachers and students.
Keywords:
Digital learning models for teachers
deep learning
graph neural networks
personalized recommendations
cognitive abilities
Journal
IF:
5.3
Papers:
2.7K
Citations:
9.0K
