On incrementally using a small portion of strong unlabeled data for semi-supervised learning algorithms.
Thanh-Binh LeSang-Woon KimPublished in: Pattern Recognit. Lett. (2014)
Keyphrases
- unlabeled data
- semi supervised learning algorithms
- semi supervised learning
- semi supervised
- labeled data
- co training
- semi supervised classification
- abundant unlabeled data
- active learning
- label propagation
- labeled training data
- supervised learning
- labeled and unlabeled data
- training data
- text classification
- multi view learning
- semi supervised learning methods
- data points
- labeled examples
- unsupervised learning
- supervised methods
- machine learning
- text categorization
- training examples
- supervised learning algorithms
- learning algorithm
- small number
- learning models
- learning problems
- pairwise constraints
- single view
- class labels
- labeled instances
- training set
- prior knowledge