PoCoTo - an Open Source System for Efficient Interactive Postcorrection of OCRed Historical Texts

Titel

PoCoTo - an Open Source System for Efficient Interactive Postcorrection of OCRed Historical Texts

DATeCH '14

Urheber

Thorsten Vobl

Annette Gotscharek

Uli Reffle

Christoph Ringlstetter

Klaus U. Schulz

Thema

decision support

error correction

user interfaces

Zusammenfassung

When applied to historical texts, OCR engines often produce a non-negligible number of OCR errors. For research in the Humanities, text mining and retrieval, the option is important to improve the quality of OCRed historical texts using interactive postcorrection. We describe a system for interactive postcorrection of OCRed historical documents developed in the EU project IMPACT. Various advanced features of the system help to efficiently correct texts. Language technology used in the background takes orthographic variation in historical language into account. Using this knowledge, the tool visualizes possible OCR errors and series of similar possible OCR errors in a given input document. Error series can be corrected in one shot. Practical user tests in three major European libraries have shown that the system considerably reduces the time needed by human correctors to eliminate a certain number of OCR errors. The system has been published as an open source tool under GitHub.

Datum

2014

Ist Teil von

Proceedings of the First International Conference on Digital Access to Textual Cultural Heritage

Verleger

New York, NY, USA

ACM

Seiten

57–61

doi

10.1145/2595188.2595197

isbn

978-1-4503-2588-2

uri

http://doi.acm.org/10.1145/2595188.2595197