Active Learning from Positive and Unlabeled Data.
Alireza GhasemiHamid R. RabieeMohsen FadaeeMohammad Taghi ManzuriMohammad H. RohbanPublished in: ICDM Workshops (2011)
Keyphrases
- active learning
- machine learning
- learning strategies
- random sampling
- training examples
- selective sampling
- training set
- supervised learning
- semi supervised
- experimental design
- annotation effort
- labeled data
- learning algorithm
- real world
- active learning strategies
- class imbalance
- imbalanced data classification
- music emotion classification
- semi supervised learning
- generalization error
- unlabeled data
- sample complexity
- batch mode
- relevance feedback
- crowd sourced
- learning process
- multiscale
- website
- data sets