Classifying OCR Errors for Use in Retrieval Methods for Norwegian Text
Master thesis
Permanent lenke
http://hdl.handle.net/11250/2615887Utgivelsesdato
2018Metadata
Vis full innførselSamlinger
Sammendrag
Analysing Norwegian documents processed with Optical Character Recognition. Using information gathered about errors, a retrieval system is tested to improve information retrieval despite OCR errors in the text. Corpus gathered from the National Library of Norway.