Posterior Distribution Learning (PDL): A novel supervised learning framework using unlabeled samples to improve classification performance.
Enmei TuJie YangNikola K. KasabovYaqian ZhangPublished in: Neurocomputing (2015)
Keyphrases
- supervised learning
- unsupervised learning
- semi supervised
- learning algorithm
- posterior distribution
- semi supervised learning
- active learning
- learning tasks
- supervised classification
- reinforcement learning
- bayesian framework
- training samples
- incremental learning
- unlabeled samples
- classification accuracy
- training set
- labeled and unlabeled data
- labeled data
- training data
- class labels
- prior knowledge
- multiple instance learning
- learning process
- posterior probability
- training examples
- text mining
- pairwise
- parameter estimation
- unlabeled data
- machine learning
- co training
- feature set
- multiple features
- support vector machine