DILBERT: Customized Pre-Training for Domain Adaptation withCategory Shift, with an Application to Aspect Extraction.
Entony LekhtmanYftah ZiserRoi ReichartPublished in: CoRR (2021)
Keyphrases
- domain adaptation
- covariate shift
- multiple sources
- cross domain
- labeled data
- semi supervised learning
- transfer learning
- sentiment classification
- training set
- semi supervised
- supervised learning
- training process
- information extraction
- unlabeled data
- target domain
- manifold alignment
- sufficient training data
- document classification
- sentiment analysis
- test data
- training examples
- model selection
- knowledge management
- training and test data
- learning environment
- neural network
- data sets