Improving Graph Neural Networks by combining active learning with self-training.
Georgios KatsimprasGeorgios PaliourasPublished in: Data Min. Knowl. Discov. (2024)
Keyphrases
- active learning
- neural network
- uncertainty sampling
- semi supervised learning
- semi supervised
- training set
- cost sensitive
- semi supervised classification
- random walk
- graph structure
- pattern recognition
- learning algorithm
- label propagation
- graph mining
- genetic algorithm
- graph model
- directed graph
- artificial neural networks
- structured data
- selective sampling
- graph theoretic
- multi layer
- weighted graph
- multilayer perceptron
- neural network model
- unlabeled data
- training examples
- labeled data
- relevance feedback
- fuzzy logic
- multi class
- graph databases
- connected components
- learning strategies
- feature space
- social networks