Jingwei Zhang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jingwei Zhang has a total of 95 co-authors. Here are the top co-authors by number of shared publications:
- Qing Yang
Guangxi Key Laboratory of Automatic Detecting Technology and Instruments, Guilin University of Electronic Technology, Guilin, Guangxi, China
9 shared publications - Qing Yang
Guangxi Key Laboratory of Automatic Detecting Technology and Instruments, Guilin University of Electronic Technology, Guilin, China
9 shared publications - Huibing Zhang
Guangxi Key Laboratory of Trusted Software, Guilin University of Electronic Technology, Guilin 541004, China
5 shared publications Qing Yang
Guangxi Key Laboratory of Automatic Detection Technology and Instrument, Guilin University of Electronic Technology
4 shared publications- Xianglin Wu
Guangxi Key Laboratory of Trusted Software, Guilin University of Electronic Technology, Guilin, China
4 shared publications - Zezheng Wu
Guangxi Key Laboratory of Trusted Software, Guilin University of Electronic Technology, Guilin, China
4 shared publications
Latest Publications
- 7 October 2026
- Published by Springer Nature in Multimedia Systems
- Vol. 32  (9) , 613
Correction: Multi-modal sarcasm detection with dynamic modality fusion and multi-frequency information propagation
- 7 October 2026
- Published by Springer Nature in Multimedia Systems
- Vol. 32  (9) , 613
Correction: Multi-modal sarcasm detection with dynamic modality fusion and multi-frequency information propagation
- 5 October 2026
- Published by Springer Nature in World Wide Web
- Vol. 29  (6) , 48
Discourse-guided multi-view adaptive fusion for multimodal emotion recognition in conversation
- 5 October 2026
- Published by Springer Nature in World Wide Web
- Vol. 29  (6) , 48
Discourse-guided multi-view adaptive fusion for multimodal emotion recognition in conversation
- 10 August 2026
- Published by Springer Nature in Multimedia Systems
- Vol. 32  (8) , 497
Multi-modal sarcasm detection with dynamic modality fusion and multi-frequency information propagation
- 10 August 2026
- Published by Springer Nature in Multimedia Systems
- Vol. 32  (8) , 497
Multi-modal sarcasm detection with dynamic modality fusion and multi-frequency information propagation
- 15 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 38  (8) , 5202-5214
Integrating Entropy Regulation and Dual-Objective Optimization for Personalized Exercise Recommendation
Personalized exercise recommendation systems aim to enhance learning efficiency by dynamically guiding students toward content aligned with their evolving knowledge states. Among various approaches, Reinforcement Learning (RL) has emerged as an effective framework for modeling student-environment interactions as sequential decision-making processes. However, most existing RL-based methods typically reward recommendations that target unmastered knowledge concepts.Such reward-driven strategies are prone to local optima, restricting exploration of unattempted or less familiar knowledge concepts. In addition, they often overlook key learning factors, such as forgetting dynamics and exercise difficulty, leading to suboptimal outcomes. To address these issues, we propose a novel Integrating Entropy Regulation and Dual-objective Optimization Exercise Recommendation (IERDO-ER) method. Specifically, we introduce an entropy-based function to encourage broader exploration in the exercise space. We also design an end-to-end policy network that generates candidate exercises as next-step actions for the recommendation agent. To enable more adaptive and pedagogically sound recommendations, we develop a dual-objective reward mechanism that incorporates both anti-forgetting and gentleness objectives. This mechanism continuously balances policy optimization through a synergy of rewards and penalties. Experiments on three real-world educational datasets show that our approach consistently outperforms strong baselines, validating its effectiveness.
- 15 May 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 38  (8) , 5202-5214
Integrating Entropy Regulation and Dual-Objective Optimization for Personalized Exercise Recommendation
Personalized exercise recommendation systems aim to enhance learning efficiency by dynamically guiding students toward content aligned with their evolving knowledge states. Among various approaches, Reinforcement Learning (RL) has emerged as an effective framework for modeling student-environment interactions as sequential decision-making processes. However, most existing RL-based methods typically reward recommendations that target unmastered knowledge concepts.Such reward-driven strategies are prone to local optima, restricting exploration of unattempted or less familiar knowledge concepts. In addition, they often overlook key learning factors, such as forgetting dynamics and exercise difficulty, leading to suboptimal outcomes. To address these issues, we propose a novel Integrating Entropy Regulation and Dual-objective Optimization Exercise Recommendation (IERDO-ER) method. Specifically, we introduce an entropy-based function to encourage broader exploration in the exercise space. We also design an end-to-end policy network that generates candidate exercises as next-step actions for the recommendation agent. To enable more adaptive and pedagogically sound recommendations, we develop a dual-objective reward mechanism that incorporates both anti-forgetting and gentleness objectives. This mechanism continuously balances policy optimization through a synergy of rewards and penalties. Experiments on three real-world educational datasets show that our approach consistently outperforms strong baselines, validating its effectiveness.
- Open Access
- 2 July 2025
- Published by Springer Nature in Discover Computing
- Vol. 28  (1) , 1-25
Intention-aware exercise recommendation enhanced by deep forgetting modeling for e-learning
- 2 July 2025
- Published by Springer Nature in Discover Computing
- Vol. 28  (1) , 1-25
Open AccessIntention-aware exercise recommendation enhanced by deep forgetting modeling for e-learning
- Open Access
- 28 May 2025
- Published by Springer Nature in Discover Computing
- Vol. 28  (1) , 1-20
Learning intention-aware knowledge tracing for learning stage
- 28 May 2025
- Published by Springer Nature in Discover Computing
- Vol. 28  (1) , 1-20
Open AccessLearning intention-aware knowledge tracing for learning stage
- Open Access
- 16 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 357-362
MESA: A Multimodal Sentiment Analysis Model Integrating Multi-Expert Mechanism and Enhanced Semantic Attention
- 16 May 2025
- Published by Association for Computing Machinery (ACM)
- p. 357-362
Open AccessMESA: A Multimodal Sentiment Analysis Model Integrating Multi-Expert Mechanism and Enhanced Semantic Attention
- Open Access
- 25 April 2025
- Published by Springer Nature in Discover Computing
- Vol. 28  (1) , 1-18
Multi-behavior enhanced group recommendation for smart educational services
- 25 April 2025
- Published by Springer Nature in Discover Computing
- Vol. 28  (1) , 1-18
Open AccessMulti-behavior enhanced group recommendation for smart educational services
- Open Access
- 15 April 2025
- Published by MDPI AG in Information
- Vol. 16  (4) , 313
Accurately Estimate and Analyze Human Postures in Classroom Environments
Estimating human posture in crowded smart teaching environments is a fundamental technical challenge for measuring learners’ engagement levels. This work presents a model for detecting critical points in human posture using ECAv2-HRNet in crowded situations. The paper introduces a method called ECAv2Net, which combines a channel feature reinforcement method with the ECANet attention mechanism network, this innovation improves the performance of the network. Additionally, ECAv2Net is integrated into the high-resolution network HRNet to create ECAv2-HRNet. This fusion allows for the incorporation of more useful feature information without increasing the model parameters. The paper also presents a human posture dataset called GUET CLASS PICTURE, which is designed for dense scenes. The experimental results when using this dataset, as well as a public dataset, demonstrate the superior performance of the human posture estimation model based on ECAv2-HRNet proposed in this paper.
- 15 April 2025
- Published by MDPI AG in Information
- Vol. 16  (4) , 313
Open AccessAccurately Estimate and Analyze Human Postures in Classroom Environments
Estimating human posture in crowded smart teaching environments is a fundamental technical challenge for measuring learners’ engagement levels. This work presents a model for detecting critical points in human posture using ECAv2-HRNet in crowded situations. The paper introduces a method called ECAv2Net, which combines a channel feature reinforcement method with the ECANet attention mechanism network, this innovation improves the performance of the network. Additionally, ECAv2Net is integrated into the high-resolution network HRNet to create ECAv2-HRNet. This fusion allows for the incorporation of more useful feature information without increasing the model parameters. The paper also presents a human posture dataset called GUET CLASS PICTURE, which is designed for dense scenes. The experimental results when using this dataset, as well as a public dataset, demonstrate the superior performance of the human posture estimation model based on ECAv2-HRNet proposed in this paper.
- 7 April 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 279
RDF-TDAA: Optimizing RDF indexing and querying with a trie based on Directly Addressable Arrays and a path-based strategy
- 7 April 2025
- Published by Elsevier in Expert Systems with Applications
- Vol. 279
RDF-TDAA: Optimizing RDF indexing and querying with a trie based on Directly Addressable Arrays and a path-based strategy