NC-WAMKD: Neighborhood Correction Weight-Adaptive Multi-Teacher Knowledge Distillation for Graph-Based Semi-Supervised Node Classification.
Jiahao LiuPengcheng GuoYonghong SongPublished in: ICASSP (2023)
Keyphrases
- semi supervised
- supervised learning
- semi supervised classification
- labeled data
- semi supervised learning algorithms
- support vector
- semi supervised learning
- co training
- classification accuracy
- labeled and unlabeled data
- knowledge sharing
- unlabeled data
- domain knowledge
- prior knowledge
- knowledge management
- supervised classification
- unsupervised learning
- knowledge representation
- active learning
- subject matter
- expert systems
- knowledge acquisition
- unlabeled samples
- decision trees
- feature weights
- classification algorithm
- label information
- classification method
- semi supervised dimensionality reduction
- semi supervised learning methods
- background knowledge
- class labels
- text classification
- pattern recognition
- feature extraction
- manifold learning
- image classification
- feature set
- support vector machine
- knowledge discovery
- knowledge base
- e learning
- feature selection
- machine learning