Yuhui Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yuhui Xu has a total of 9 co-authors. Here are the top co-authors by number of shared publications:
- Qin Ni
Shanghai International Studies University
2 shared publications - Bo Zhang
Shanghai Normal University
1 shared publication Liang He
Shanghai Key Laboratory of Multidimensional information Processing, School of Computer Science and Technology, East China Normal University, Shanghai, China
1 shared publication- Shuang Liu
College of Information, Mechanical and Electrical Engineering, Shanghai Normal University, Shanghai, China
1 shared publication - Yangze Yu
Shanghai Engineering Research Center of Intelligent Education and Bigdata, Shanghai Normal University, Fengxian, Shanghai, China
1 shared publication - Yifei Mi
Lab for Educational Big Data and Policymaking, Shanghai Normal University, Fengxian, Shanghai, China
1 shared publication
Latest Publications
- 24 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Learning Technologies
- Vol. 17, 241-257
Design and Implementation of the Reliable Learning Style Recognition Mechanism Based on Fusion Labels and Ensemble Classification
Learning style recognition is an indispensable part of achieving personalized learning in online learning systems. The traditional inventory method for learning style identification faces the limitations such as subject and static characteristics. Therefore, an automatic and reliable learning style recognition mechanism is designed in this paper. Firstly, a learning style labeling framework (LSDFA) based on multi-label fusion is proposed, which can obtain learning style labels by mining the potential information of two sets of inventories. Furthermore, a two-layer ensemble model (SRGSML) based on learners' online learning behaviors data to recognize learners' learning styles is proposed, which combines the resampling technology (SMOTE) to solve the unreliable prediction problem caused by class imbalance. The superiority of the proposed mechanism is verified on learning behavior data of 2,056 learners during the online teaching period of Shanghai Normal University. Experimental results show that the recognition accuracy of SRGSML achieves to 0.977, as well as prove the effectiveness of the LSDFA for labeling learning style.
- 24 April 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Learning Technologies
- Vol. 17, 241-257
Design and Implementation of the Reliable Learning Style Recognition Mechanism Based on Fusion Labels and Ensemble Classification
Learning style recognition is an indispensable part of achieving personalized learning in online learning systems. The traditional inventory method for learning style identification faces the limitations such as subject and static characteristics. Therefore, an automatic and reliable learning style recognition mechanism is designed in this paper. Firstly, a learning style labeling framework (LSDFA) based on multi-label fusion is proposed, which can obtain learning style labels by mining the potential information of two sets of inventories. Furthermore, a two-layer ensemble model (SRGSML) based on learners' online learning behaviors data to recognize learners' learning styles is proposed, which combines the resampling technology (SMOTE) to solve the unreliable prediction problem caused by class imbalance. The superiority of the proposed mechanism is verified on learning behavior data of 2,056 learners during the online teaching period of Shanghai Normal University. Experimental results show that the recognition accuracy of SRGSML achieves to 0.977, as well as prove the effectiveness of the LSDFA for labeling learning style.
- Open Access
- 27 December 2022
- Published by Springer Nature in International Journal of Computational Intelligence Systems
- Vol. 15 (1) , 1-22
Learning Style Integrated Deep Reinforcement Learning Framework for Programming Problem Recommendation in Online Judge System
Exercise recommendation is an integral part of enabling personalized learning. Giving appropriate exercises can facilitate learning for learners. The programming problem recommendation is a specific application of the exercise recommendation. Therefore, an innovative recommendation framework for programming problems that integrate learners’ learning styles is proposed. In addition, there are some difficulties to be solved in this framework, such as quantifying learning behavior, representing programming problems, and quantifying learning strategies. For the difficulties in quantifying learning behavior and quantifying learning strategies, a programming problem recommendation algorithm based on deep reinforcement learning (DRLP) is proposed. DRLP includes the specific design of action space, action-value Q-network, and reward function. Learning style is embedded into DRLP through action space to make recommendations more personalized. To represent the programming problem in DRLP, a multi-dimensional integrated programming problem representation model is proposed to quantify the difficulty feature, knowledge point feature, text description, input description, and output description of programming problems. In particular, Bi-GRU is introduced to learn texts’ contextual semantic association information from both positive and negative directions. Finally, a simulation experiment is carried out with the actual learning behavior data of 47,147 learners in the LUOGU Online Judge system. Compared with the optimal baseline model, the recommendation effect of DRLP has improved (HR, MRR, and Novelty have increased by 4.35%, 1.15%, and 1.1%), which proves the rationality of the programming problem representation model and action-value Q-network.
- 27 December 2022
- Published by Springer Nature in International Journal of Computational Intelligence Systems
- Vol. 15 (1) , 1-22
Open AccessLearning Style Integrated Deep Reinforcement Learning Framework for Programming Problem Recommendation in Online Judge System
Exercise recommendation is an integral part of enabling personalized learning. Giving appropriate exercises can facilitate learning for learners. The programming problem recommendation is a specific application of the exercise recommendation. Therefore, an innovative recommendation framework for programming problems that integrate learners’ learning styles is proposed. In addition, there are some difficulties to be solved in this framework, such as quantifying learning behavior, representing programming problems, and quantifying learning strategies. For the difficulties in quantifying learning behavior and quantifying learning strategies, a programming problem recommendation algorithm based on deep reinforcement learning (DRLP) is proposed. DRLP includes the specific design of action space, action-value Q-network, and reward function. Learning style is embedded into DRLP through action space to make recommendations more personalized. To represent the programming problem in DRLP, a multi-dimensional integrated programming problem representation model is proposed to quantify the difficulty feature, knowledge point feature, text description, input description, and output description of programming problems. In particular, Bi-GRU is introduced to learn texts’ contextual semantic association information from both positive and negative directions. Finally, a simulation experiment is carried out with the actual learning behavior data of 47,147 learners in the LUOGU Online Judge system. Compared with the optimal baseline model, the recommendation effect of DRLP has improved (HR, MRR, and Novelty have increased by 4.35%, 1.15%, and 1.1%), which proves the rationality of the programming problem representation model and action-value Q-network.