Introduction to the special issue on computational linguistics using large corpora
Computational Linguistics - Special issue on using large corpora: I
Hi-index | 0.00 |
According to the statistic analysis of common errors found in those texts that are typed-in, OCR-recognized or phonetics-recognized but not be proofread and the characteristics of such texts, we propose a error-detecting principle and error-detecting algorithm based on the orderly-neighborship. Furthermore, Factors that affect performance index of error-detecting system such as recall ratio and accurate ratio are discussed.