Please use this identifier to cite or link to this item:
https://scholarbank.nus.edu.sg/handle/10635/40544
DC Field | Value | |
---|---|---|
dc.title | Influence of language models and candidate set size on contextual post-processing for chinese script recognition | |
dc.contributor.author | Li, Y.-X. | |
dc.contributor.author | Tan, C.L. | |
dc.date.accessioned | 2013-07-04T08:06:47Z | |
dc.date.available | 2013-07-04T08:06:47Z | |
dc.date.issued | 2004 | |
dc.identifier.citation | Li, Y.-X.,Tan, C.L. (2004). Influence of language models and candidate set size on contextual post-processing for chinese script recognition. Proceedings - International Conference on Pattern Recognition 2 : 537-540. ScholarBank@NUS Repository. | |
dc.identifier.issn | 10514651 | |
dc.identifier.uri | http://scholarbank.nus.edu.sg/handle/10635/40544 | |
dc.description.abstract | In the Chinese language, a word consisting of one or more characters is a basic syntax-meaningful unit, however, each character in the word also has a definite meaning in itself. In this paper, we compare the perplexities of four n-gram language models (character-based bigram, character-based trigram, word-based bigram and class-based bigram) and their influence on the performance of contextual post-processing of Chinese scripts in an offline handwritten Chinese character recognition system. We also demonstrate the influence of the candidate set size on the performance of contextual post-processing in detail, and indicate that the number of candidates should vary with each script. | |
dc.source | Scopus | |
dc.type | Conference Paper | |
dc.contributor.department | COMPUTER SCIENCE | |
dc.description.sourcetitle | Proceedings - International Conference on Pattern Recognition | |
dc.description.volume | 2 | |
dc.description.page | 537-540 | |
dc.description.coden | PICRE | |
dc.identifier.isiut | NOT_IN_WOS | |
Appears in Collections: | Staff Publications |
Show simple item record
Files in This Item:
There are no files associated with this item.
Google ScholarTM
Check
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.