Estimating the class prior and posterior from noisy positives and unlabeled data.
Shantanu JainMartha WhitePredrag RadivojacPublished in: NIPS (2016)
Keyphrases
- unlabeled data
- labeled data
- semi supervised learning
- class labels
- semi supervised
- class distribution
- active learning
- prior knowledge
- co training
- label information
- positive examples
- training data
- semi supervised classification
- text classification
- labeled and unlabeled data
- learning algorithm
- supervised learning
- labeled examples
- data points
- text categorization
- labeled training data
- domain adaptation
- training set
- transfer learning
- data sets
- number of labeled examples
- gaussian process
- bayesian framework
- label propagation
- machine learning
- cost sensitive
- maximum a posteriori
- unsupervised learning
- multi view
- high dimensional
- object recognition
- supervised learning algorithms
- unlabeled instances
- data mining