DocumentCode
3647682
Title
Turkish archive digitization by human computation approach
Author
İbrahim Gümüş;Osman Abul
Author_Institution
Department of Computer Engineering, TOBB University of Economics and Technology, Ankara, Turkey
fYear
2012
fDate
7/1/2012 12:00:00 AM
Firstpage
1
Lastpage
4
Abstract
Human computation is a promising approach at solving computation tasks for which humans currently perform better than computers. Archive digitization is such a task. In this paper, we present our reCAPTCHA like archive digitization system, called trCAPTCHA, mainly targeting old Turkish archives and hence Turkish speaking audience. trCAPTCHA differs from reCAPTCHA that it uses no global dictionary but constructs a local dictionary for each scanned word from alternative texts generated through OCR readings.
Keywords
"Humans","Optical character recognition software","Computers","Dictionaries","Games","Databases","Accuracy"
Publisher
ieee
Conference_Titel
Innovations in Intelligent Systems and Applications (INISTA), 2012 International Symposium on
Print_ISBN
978-1-4673-1446-6
Type
conf
DOI
10.1109/INISTA.2012.6247001
Filename
6247001
Link To Document