Abstract
In this paper we demonstrate a finite-state implementation of context-aware spell checking utilizing an N-gram based part of speech (POS) tagger to rerank the suggestions from a simple edit-distance based spell-checker. We demonstrate the benefits of context-aware spell-checking for English and Finnish and introduce modifications that are necessary to make traditional N-gram models work for morphologically more complex languages, such as Finnish.
| Original language | English |
|---|---|
| Title of host publication | Computational Linguistics and Intelligent Text Processing : 13th International Conference, CICLing 2012 |
| Editors | Alexander Gelbukh |
| Number of pages | 11 |
| Place of Publication | Delhi, India |
| Publication date | 9 Mar 2012 |
| Publication status | Published - 9 Mar 2012 |
| MoE publication type | A4 Article in conference proceedings |
| Event | International Conference on Intelligent Text Processing and Computational Linguistics - New Delhi, India Duration: 11 Mar 2012 → 17 Mar 2012 |
Fields of Science
- 6121 Languages
- Language technology
- HFST
- spelling suggestions