Semi-supervised learning using higher-order co-occurrence paths to overcome the complexity of data representation.
Murat Can GanizPublished in: SMC (2016)
Keyphrases
- co occurrence
- semi supervised learning
- data representation
- higher order
- semi supervised
- unlabeled data
- labeled data
- data representations
- supervised learning
- machine learning
- dimensionality reduction
- wordnet
- unsupervised learning
- active learning
- text categorization
- learning models
- training data
- high order
- label propagation
- pairwise
- conditional random fields
- named entities
- pairwise constraints
- xml documents
- domain specific
- decision trees
- topic models
- xml schema
- knowledge base
- probabilistic latent semantic analysis