Grammatical error correction using hybrid systems and type filtering
- Submitting institution
-
University of Cambridge
- Unit of assessment
- 11 - Computer Science and Informatics
- Output identifier
- 7174
- Type
- E - Conference contribution
- DOI
-
10.3115/v1/w14-1702
- Title of conference / published proceedings
- CoNLL 2014 - 18th Conference on Computational Natural Language Learning, Proceedings of the Shared Task
- First page
- 15
- Volume
- -
- Issue
- -
- ISSN
- -
- Open access status
- -
- Month of publication
- January
- Year of publication
- 2014
- URL
-
-
- Supplementary information
-
-
- Request cross-referral to
- -
- Output has been delayed by COVID-19
- No
- COVID-19 affected output statement
- -
- Forensic science
- No
- Criminology
- No
- Interdisciplinary
- No
- Number of additional authors
-
4
- Research group(s)
-
-
- Citation count
- -
- Proposed double-weighted
- No
- Reserve for an output with double weighting
- No
- Additional information
- This paper describes our winning submission to the 2014 CoNLL shared task (international competition) on grammatical error correction (GEC). It was the first paper that demonstrated that a statistical machine translation (SMT) approach from 'bad' to 'good' English produced state-of-the-art results when combined with an appropriate language model to select the best correction. This spawned a large amount of further research on SMT, and subsequently neural MT, for GEC, leading fairly directly to the current generation of much improved writing assistance technology deployed, for example, by Google in Gmail.
- Author contribution statement
- -
- Non-English
- No
- English abstract
- -