Improving Grammatical Error Correction via Pre-Training a Copy-Augmented Architecture with Unlabeled Data.
Wei ZhaoLiang WangKewei ShenRuoyu JiaJingming LiuPublished in: NAACL-HLT (1) (2019)
Keyphrases
- error correction
- unlabeled data
- labeled data
- number of labeled examples
- semi supervised learning
- labeled training data
- supervised learning
- training set
- semi supervised
- labeled data for training
- training examples
- active learning
- semi supervised classification
- co training
- labeled examples
- training data
- text classification
- error detection
- training and test data
- error correcting
- labeled and unlabeled data
- class labels
- learning algorithm
- text categorization
- data points
- transfer learning
- labeled instances
- prior knowledge
- supervised learning algorithms
- domain adaptation
- watermarking scheme
- unlabeled training data
- data sets
- positive examples
- machine learning
- unsupervised learning
- ensemble learning
- support vector machine
- training process
- text mining
- text classifiers
- training samples
- data mining