A5042460236- Advanced Graph Neural Networks
- Recommender Systems and Techniques
- Sentiment Analysis and Opinion Mining
- Topic Modeling
- Advanced Text Analysis Techniques
- Advanced Image and Video Retrieval Techniques
- Caching and Content Delivery
- Music and Audio Processing
- Domain Adaptation and Few-Shot Learning
- Misinformation and Its Impacts
- Complex Network Analysis Techniques
- Face recognition and analysis
- Generative Adversarial Networks and Image Synthesis
- Speech and Audio Processing
- Text and Document Classification Technologies
- Video Analysis and Summarization
- Advanced Image Processing Techniques
- Advanced Data and IoT Technologies
- Distributed systems and fault tolerance
- Pneumocystis jirovecii pneumonia detection and treatment
- Diabetic Foot Ulcer Assessment and Management
- Expert finding and Q&A systems
- Remote-Sensing Image Classification
- Graph Labeling and Dimension Problems
- Spam and Phishing Detection
Central China Normal University
2019-2024
Shenzhen Third People’s Hospital
2022
Soochow University
2021
Wuhan National Laboratory for Optoelectronics
2017-2019
Huazhong University of Science and Technology
2016-2019
Aspect-level sentiment classification is an interesting but challenging research problem, namely, the prediction of polarity toward a specific aspect term opinionated sentence. Previous attention-based recurrent neural networks have been proposed to address this problem because attention mechanism capable finding out those words contributing more than others and shown great promise. However, major drawback these approaches that explicit position context ignored. Drawing inspirations from...
Abstract Diabetic ulcers (DUs) appearing as chronic wounds are difficult to heal due the oxidative stress in wound microenvironment and their high susceptibility bacterial infection. A routine treatment combining surgical debridement with anti‐infection therapy is widely used for treating DUs clinic, but hardly offers a satisfying healing outcome. It known that long‐term antibiotic may also lead drug resistance of pathogens. To address these challenges, new strategies both reactive oxygen...
As a classical technique in storage systems, disk failure prediction aims at predicting impending failures advance for high data reliability. Over the past decades, taking as input SMART (Self-Monitoring, Analysis and Reporting Technology) attributes, many supervised machine learning algorithms have been proven to be effective prediction. However, these approaches heavily rely on availability of substantial annotated failed which unfortunately exhibits an extreme imbalance, i.e., number...
Face age progression/regression has garnered substantial active research interest due to its tremendous impact on a wide-range of practical applications like searching for missing individuals with photos childhood, entertainment, and so on. Most existing face aging models have proven be successful effective in learning the transformation between groups aid paired samples, i.e., images same person at different ages. Considering expensive cost collecting datasets, Conditional Adversarial...
Human perception heavily relies on two primary senses: vision and hearing, which are closely inter-connected capable of complementing each other. Consequently, various multimodal learning tasks have emerged, with audio-visual event localization (AVEL) being a prominent example. AVEL is popular task within the realm learning, objective identifying presence events video segment predicting their respective categories. This holds significant utility in domains such as healthcare monitoring...
Many online social networks can be described by signed networks, where positive links signify friendships, trust and like; while negative indicate enmity, distrust dislike. Predicting the sign of in these has attracted a great deal attentions areas friendship recommendation relationship prediction. Existing methods for prediction tend to rely on path-based features which are somehow limited sparsity problem network. In order solve this issue, paper, we introduce novel model exploiting...
There has been growing attention on explainable recommendation that is able to provide high-quality results as well intuitive explanations. However, most existing studies use offline prediction strategies where recommender systems are trained once while used forever, which ignores the dynamic and evolving nature of user–item interactions. two main issues with these methods. First, their random dataset split setting will result in data leakage knowledge should not be known at time training...
Abstract Background Pre-extensively drug-resistant tuberculosis (Pre XDR-TB) was defined as resistant to fluroquinolones in Multidrug (MDR TB), it had poorer outcomes than MDR TB previous reports. In this study, we aimed evaluate the efficacy and safety of Bdq containing regimen for treatment Pre XDR-TB. Moreover, tried explore optimal duration total duration. Patients methods : retrospective a 84 XDR-TB from our hospital were enrolled divided into group (46 cases) non (38 according their...