Japanese OCR Error Correction Using Stochastic Morphological Analyzer and Probabilistic Word N-gram Model

2000 ◽  
Vol 13 (01) ◽  
pp. 69-82 ◽  
Author(s):  
KOICHI TAKEUCHI ◽  
YUJI MATSUMOTO
Keyword(s):  
2014 ◽  
Vol 41 (12) ◽  
pp. 1081-1089 ◽  
Author(s):  
Minho Kim ◽  
Hyuk-Chul Kwon ◽  
Sungki Choi

Author(s):  
Xurong Xu

To discuss the grammar checking system and to solve the low accuracy of the system and bad effect, the grammar checking algorithm and the accuracy were improved. On this basis, a grammar checking system was designed and developed. The grammar checking module adopted the grammar checking design based on multi-rules and used multi-layer grammar rules to provide error correction function for common English grammar. To improve the grammar checking accuracy of grammar checking module, the grammatical model based on statistics was introduced by the system. In addition, N-gram model was applied to evaluate the errors found out by the rules grammar. The experimental results showed that the grammar checking system carried out grammar check of domestic students’ English composition samples, and the system performance was good. It is concluded that in asynchronous mode, hundreds of sentences can be checked at the same time, and the purpose of the design is achieved.


Sign in / Sign up

Export Citation Format

Share Document