| LREC 2000 2nd International Conference on Language Resources & Evaluation | ||||||
|
|
|
|
|
|
|
| Title | Typographical and Orthographical Spelling Error Correction |
| Authors | Min Kyongho (School of Computer Science and Engineering, The University of New South Wales, Sydney NSW 2052, Australia, min@cse.unsw.edu.au) Wilson William H. (School of Computer Science and Engineering, The University of New South Wales, Sydney NSW 2052, Australia, billw@cse.unsw.edu.au) Moon Yoo-Jin (Department of Management Information System, Hannam University, Ojung-dong, Daeduk-ku, Daejun, 300-791, Korea, yjmoon@eve.hannam.ac.kr) |
| Keywords | |
| Session | Session WP9 - Applications using Written Language Resources |
| Full Paper | 221.ps, 221.pdf |
| Abstract | This paper focuses on selection techniques for best correction of misspelt words at the lexical level. Spelling errors are introduced by either cognitive or typographical mistakes. A robust spelling correction algorithm is needed to cover both cognitive and typographical errors. For the most effective spelling correction system, various strategies are considered in this paper: ranking heuristics, correction algorithms, and correction priority strategies for the best selection. The strategies also take account of error types, syntactic information, word frequency statistics, and character distance. The findings show that it is very hard to generalise the spelling correction strategy for various types of data sets such as typographical, orthographical, and scanning errors. |