Mitigating shortage of labeled data using clustering-based active learning with diversity exploration.
Xuyang YanShabnam NazmiBiniam GebruMohd AnwarAbdollah HomaifarMrinmoy SarkarKishor Datta GuptaPublished in: CoRR (2022)
Keyphrases
- labeled data
- active learning
- unlabeled data
- semi supervised learning
- semi supervised
- transfer learning
- supervised learning
- unlabeled examples
- labeled training data
- training examples
- labeled examples
- unlabeled instances
- class labels
- semi supervised classification
- learning algorithm
- training set
- domain adaptation
- training data
- fully supervised
- selective sampling
- labeled instances
- prior knowledge
- learning process
- labeled and unlabeled data
- data mining
- label information
- co training
- text classification
- machine learning
- data sets
- random sampling
- unsupervised learning
- labeling effort
- labeled training set
- labeled data for training