Abstract
This paper presents a standalone spell corrector, WNSpell, based on and written for WordNet. It is aimed at generating the best possible suggestion for a mistyped query but can also serve as an all-purpose spell corrector. The spell corrector consists of a standard initial correction system, which evaluates word entries using a multifaceted approach to achieve the best results, and a semantic recognition system, wherein given a related word input, the system will adjust the spelling suggestions accordingly. Both feature significant performance improvements over current context-free spell correctors.- Anthology ID:
- 2016.gwc-1.21
- Volume:
- Proceedings of the 8th Global WordNet Conference (GWC)
- Month:
- 27--30 January
- Year:
- 2016
- Address:
- Bucharest, Romania
- Editors:
- Christiane Fellbaum, Piek Vossen, Verginica Barbu Mititelu, Corina Forascu
- Venue:
- GWC
- SIG:
- SIGLEX
- Publisher:
- Global Wordnet Association
- Note:
- Pages:
- 136–143
- Language:
- URL:
- https://aclanthology.org/2016.gwc-1.21
- DOI:
- Cite (ACL):
- Bill Huang. 2016. WNSpell: a WordNet-Based Spell Corrector. In Proceedings of the 8th Global WordNet Conference (GWC), pages 136–143, Bucharest, Romania. Global Wordnet Association.
- Cite (Informal):
- WNSpell: a WordNet-Based Spell Corrector (Huang, GWC 2016)
- PDF:
- https://preview.aclanthology.org/ml4al-ingestion/2016.gwc-1.21.pdf