A5042246608- Handwritten Text Recognition Techniques
- Data Stream Mining Techniques
- Domain Adaptation and Few-Shot Learning
- Online Learning and Analytics
- Text and Document Classification Technologies
- Advanced Neural Network Applications
- Image Processing and 3D Reconstruction
- Natural Language Processing Techniques
- Machine Learning and ELM
- Forensic and Genetic Research
- Digital Media Forensic Detection
- Machine Learning in Healthcare
- Topic Modeling
Shanghai University of Engineering Science
2024
Beijing University of Posts and Telecommunications
2022
Baotou Teachers College
2022
Student Dropout Prediction (SDP) is pivotal in mitigating withdrawals Massive Open Online Courses. Previous studies generally modeled the SDP problem as a binary classification task, providing single prediction outcome. Accordingly, some attempts introduce survival analysis methods to achieve continuous and consistent predictions over time. However, volatility sparsity of data always weaken models' performance. Prevailing solutions rely heavily on pre-processing independent predictive...
Oraclebone characters (OBCs) are crucial for understanding ancient Chinese history, but existing recognition methods only recognize known categories in labeled data, neglecting novel unlabeled data. This work introduces a approach to discovering new OBC data through generalized category discovery. We address the challenges posed by OBCs’ instinctive characteristics, such as misleading contrastive views from random cropping, sub-optimal learned representation, and insufficient supervision Our...
Student Dropout Prediction (SDP) is of pivotal significance in mitigating withdrawals Massive Open Online Courses. Research these areas are usually carried out using deep learning to detect complex nonlinear patterns students' sequences. However, the volatility and sparsity data always weaken performance neural networks. Prevailing approaches required an additional smoothing or interpolation step independent prediction model, which may lose valuable information introduce inauthentic data....