Please use this identifier to cite or link to this item: http://hdl.handle.net/10397/31341
Title: A multiple classifier approach to detect Chinese character recognition errors
Authors: Hung, KY
Luk, RWP 
Yeung, DS
Chung, KFL 
Shu, W
Keywords: Character recognition
Error detection
Pattern recognition and language modeling
Issue Date: 2005
Publisher: Elsevier
Source: Pattern recognition, 2005, v. 38, no. 5, p. 723-738 How to cite?
Journal: Pattern recognition 
Abstract: Detection of recognition errors is important in many areas, such as improving recognition performance, saving manual effort for proof-reading and post-editing, and assigning appropriate weights for retrieval in constructing digital libraries. We propose a novel application of multiple classifiers for the detection of recognition errors. A need for multiple classifiers emerges when a single classifier cannot improve recognition-error detection performance compared with the current detection scheme using a simple threshold mechanism. Although the single classifier does not improve recognition error performance, it serves as a baseline for comparison and the related study of useful features for error detection suggests three distinct cases where improvement is needed. For each case, the multiple classifier approach assigns a classifier to detect the presence or absence of errors and additional features are considered for each case. Our results show that the recall rate (70-80%) of recognition errors, the precision rate (80-90%) of recognition error detection and the saving in manual effort (75%) were better than the corresponding performance using a single classifier or a simple threshold detection scheme.
URI: http://hdl.handle.net/10397/31341
ISSN: 0031-3203
EISSN: 1873-5142
DOI: 10.1016/j.patcog.2004.09.004
Appears in Collections:Journal/Magazine Article

Access
View full-text via PolyU eLinks SFX Query
Show full item record

SCOPUSTM   
Citations

4
Last Week
0
Last month
0
Citations as of Sep 11, 2017

WEB OF SCIENCETM
Citations

1
Last Week
0
Last month
0
Citations as of Sep 14, 2017

Page view(s)

43
Last Week
3
Last month
Checked on Sep 18, 2017

Google ScholarTM

Check

Altmetric



Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.