A5100353514- Online Learning and Analytics
- Online and Blended Learning
- Energy Load and Power Forecasting
- Advanced Wireless Communication Techniques
- Time Series Analysis and Forecasting
- Coding theory and cryptography
- Wireless Communication Networks Research
- Hydrological Forecasting Using AI
- Intelligent Tutoring Systems and Adaptive Learning
- Video Analysis and Summarization
- Seismology and Earthquake Studies
- Data Stream Mining Techniques
- Multimodal Machine Learning Applications
- Anomaly Detection Techniques and Applications
- Flood Risk Assessment and Management
- ECG Monitoring and Analysis
- Advanced Image and Video Retrieval Techniques
- Mobile Crowdsensing and Crowdsourcing
- EEG and Brain-Computer Interfaces
Central China Normal University
2007-2023
NetEase (China)
2023
In a massive open online courses (MOOCs) learning environment, it is essential to understand students' social knowledge constructs and critical thinking for instructors design intervention strategies. The development of can be represented by cognitive presence, which primary component the community inquiry model. However, identifying learners' presence challenging problem, most researchers have performed this task using traditional machine methods that require both manual feature...
With the continuous development of online learning platforms, educational data analytics and prediction have become a promising research field, which are helpful for personalized system. However, indicator's selection process does not combine with whole process, may affect accuracy results. In this paper, we induce 19 behavior indicators in platform, proposing student performance model combines process. The consists four parts: collection pre-processing, analytics, algorithm building...
In order to further improve the accuracy of rainfall prediction, this paper proposes a Bi-LSTM prediction model incorporating an attention mechanism. The splits input sequence into spatio-temporal and feature sequence, introduces mechanism before network calculate two sequences separately fusion. can adaptively select most important features according their importance, parameters layer are obtained by competitive random search algorithm, which enhances robustness model. Finally, experiments...