Improving State-of-the-Art in One-Class Classification by Leveraging Unlabeled Data.
Farid BagirovDmitry IvanovAleksei ShpilmanPublished in: CoRR (2022)
Keyphrases
- unlabeled data
- labeled data
- semi supervised learning
- semi supervised
- co training
- semi supervised classification
- active learning
- training data
- supervised learning
- labeled and unlabeled data
- learning algorithm
- training set
- labeled examples
- text classification
- data points
- number of labeled examples
- text categorization
- class distribution
- supervised and unsupervised learning
- training examples
- class labels
- domain adaptation
- labeled training data
- small set of labeled
- label propagation
- learning problems
- prior knowledge
- positive examples
- information retrieval
- multi view learning
- semisupervised learning
- binary classification
- unsupervised learning
- supervised learning algorithms
- high dimensional
- pairwise
- labeled data for training
- select relevant features