nad a22 4500 555163199 20190202120452.0 cr unu---uuuuu 190202s2018 xx s 000 0 eng 10.5167/uzh-162394 doi (ZORA)oai:www.zora.uzh.ch:162394 000 ddc 410 ddc Amrhein Chantal Supervised OCR Error Detection and Correction Using Statistical and Neural Machine Translation Methods [Elektronische Daten] [Chantal Amrhein, Simon Clematide] openAccess eu-repo For indexing the content of digitized historical texts, optical character recognition (OCR) errors are a hampering problem. To explore the effectivity of new strategies for OCR post-correction, this article focuses on methods of character-based machine translation, specifically neural machine translation and statistical machine translation. Using the ICDAR 2017 data set on OCR post-correction for English and French, we experiment with different strategies for error detection and error correction. We analyze how OCR post-correction with NMT can profit from using additional information and show that SMT and NMT can benefit from each other for these tasks. An ensemble of our models reached best performance in ICDAR's 2017 error correction subtask and performed competitively in error detection. However, our experimental results also suggest that tuning supervised learning for OCR post-correction of texts from different sources, text types (periodicals and monographs), time periods and languages is a difficult task: the data on which the MT systems are trained have a large influence on which methods and features work best. Conclusive and generally applicable insights are hard to achieve. Institute of Computational Linguistics zora OCR post-correction Machine Learning Neural Machine Translation Statistical Machine Translation zora Clematide Simon joint author Journal for Language Technology and Computational Linguistics (JLCL) 33(1):49-76 https://www.zora.uzh.ch/id/eprint/162394/1/AmrheinClematide2018.pdf text/html WWW-Backlink auf das Repository Onlinezugriff via WWW text/html https://jlcl.org/content/2-allissues/1-heft1-2018/jlcl_2018-1_3.pdf ZORA 1 Journal Article PeerReviewed zora SNSF/Projectfunding/CRSII5_173719/CH zora grantAgreement ZORA 856 40 https://www.zora.uzh.ch/id/eprint/162394/1/AmrheinClematide2018.pdf text/html WWW-Backlink auf das Repository ZORA 856 42 Onlinezugriff via WWW text/html https://jlcl.org/content/2-allissues/1-heft1-2018/jlcl_2018-1_3.pdf ZORA 100 1- Amrhein Chantal ZORA 700 1- Clematide Simon joint author ZORA 773 0- Journal for Language Technology and Computational Linguistics (JLCL) 33(1):49-76 BK010053 XK010053 XK010000 ZORA ZORA ZORA Journal Article openAccess