A5011073406- Advanced Graph Neural Networks
- Complex Network Analysis Techniques
- Topic Modeling
- Graph Theory and Algorithms
- Recommender Systems and Techniques
- Text and Document Classification Technologies
- Concrete and Cement Materials Research
- Magnesium Oxide Properties and Applications
- Neural Networks and Applications
- Sentiment Analysis and Opinion Mining
- Data Quality and Management
- Concrete Corrosion and Durability
- Bioinformatics and Genomic Networks
Fuzhou University
2022-2024
Chengdu University of Technology
2024
Recent years have witnessed a drastic surge in graph representation learning, which usually produces low-dimensional and crisp representations from topology high-dimensional node attributes. Nevertheless, of or actually conceals the uncertainty interpretability features. In citation networks, for example, reference between two papers is always involved with fuzziness denoting correlation degrees, that is, one connection may simultaneously belong to strong weak references different beliefs....
A large number of real-world systems generate graphs that are structured data aligned with nodes and edges. Graphs usually dynamic in many scenarios, where or edges keep evolving over time. Recently, graph representation learning (GRL) has received great success network analysis, which aims to produce informative representative features low-dimensional embeddings by exploring node attributes topology. Most state-of-the-art models for GRL composed a static model recurrent neural (RNN). The...
Exploring dynamic patterns from complex and large-scale networks is a significant challenging task in graph analysis. One of the most advanced solutions representation learning, which embeds structural temporal correlations into representative vector for each node or subgraph. Existing models have made some successes, such as overcoming problems induction unseen nodes scalability evolving networks. However, these usually rely on crisp learning that incapable modeling feature fuzziness...
With the rise of contrastive learning, unsupervised graph representation learning has shown strong competitiveness. However, existing models typically either focus on local view graphs or take simple considerations both global and views. This may cause these to overemphasize importance individual nodes their ego-networks, result in poor knowledge affect Additionally, most pay attention topological proximity, assuming that are closer topology more similar. real world, close be dissimilar,...