Investigating the Effectiveness of Representations Based on Pretrained Transformer-based Language Models in Active Learning for Labelling Text Datasets.
Jinghui LuBrian MacNameePublished in: CoRR (2020)
Keyphrases
- language model
- active learning
- information retrieval
- language modeling
- probabilistic model
- n gram
- language modelling
- document retrieval
- speech recognition
- word clouds
- document level
- multiword
- query expansion
- statistical language models
- test collection
- retrieval model
- text retrieval
- context sensitive
- language models for information retrieval
- document representation
- text mining
- document length
- document ranking
- pseudo relevance feedback
- search engine
- text documents
- relevance model
- query terms
- learning algorithm
- keywords
- semi supervised
- machine learning