Larramendiren "Hiztegi Hirukoitza"ren digitalizazioa. Karaktereen ezagutze optikoa eta "Wikiteka"ra igotzea
No Thumbnail Available
Date
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Artikulu honetan Larramendiren Hiztegi Hirukoitzaren digitalizazioko OCR prozesua deskribatzenda, ikasketa automatikoa baliatuz. Horretarako, irudien tratamendua deskribatu eta eskuztranskribatutako laginetik abiatuta testua ezagutuko duen ereduaren trebakuntza azaltzen da. EmaitzakWikiteka plataforman eskuragarri daudenez, auzolanaz transkripzio osoa zuzendutakoan informazioerauzketa prozesutik pasako da hiztegiaren egitura lexikografikoa ikasketa automatikoz erauzteko.
In this article, we describe the OCR process using machine learning in the digitization ofLarramendi’s Diccionario Trilingüe. For this purpose, the treatment of images is described and thetraining of the model from the transcribed sample that will recognize the text. As the results areavailable on the Wikisource platform, the transcription can be corrected using crowdsourcing, so thatwe can carry out the information extraction process using machine learning to extract thelexicographic structure of the dictionary.
In this article, we describe the OCR process using machine learning in the digitization ofLarramendi’s Diccionario Trilingüe. For this purpose, the treatment of images is described and thetraining of the model from the transcribed sample that will recognize the text. As the results areavailable on the Wikisource platform, the transcription can be corrected using crowdsourcing, so thatwe can carry out the information extraction process using machine learning to extract thelexicographic structure of the dictionary.
Description
Keywords
hiztegi historikoak, Larramendi, OCR, ikasketa automatikoa, Wikiteka, historical dictionaries, Larramendi, OCR, machine learning, Wikisource