A5104426089- Topic Modeling
- Web Data Mining and Analysis
- Advanced Graph Neural Networks
- Advanced Neural Network Applications
- Data Quality and Management
- Brain Tumor Detection and Classification
- Domain Adaptation and Few-Shot Learning
- Biomedical Text Mining and Ontologies
Shanghai Institute of Technology
2024
Lightweight, high‐performance networks are important in vision perception systems. Recent research on convolutional neural has shown that attention mechanisms can significantly improve the network performance. However, existing approaches either ignore significance of using both types (channel and space) simultaneously or increase model complexity. In this study, we propose adaptive module (AAM), which is a truly lightweight yet effective comprises channel spatial submodules to balance...
To address the limitations in capturing complex semantic features between entities and incomplete acquisition of entity relationship information by existing patent knowledge graph reasoning algorithms, we propose a method that integrates structural for graphs, denoted as SS-DSA. Initially, to facilitate model representation information, directed based on is designed. Subsequently, within mined using inductive learning, which combined with learning features. Finally, an attention mechanism...
Aimed at mitigating the limitations of existing document entity relation extraction methods, especially complex information interaction between different entities in and poor effect classification, according to semi-structured characteristics patent data, a ontology model construction method based on hierarchical clustering association rules was proposed describe their relations document, dubbed as MPreA. Combined with statistical learning deep algorithms, pre-trained attention mechanism...