Toolkit to assess OCR’ed historical text in the era of big data
LYRASIS; University of Utah
While cultural heritage institutions have been using Optical Character Recognition (OCR) to extract full text from scanned page images, the quality of extracted text is low for historical texts. In this era of big data, such historical texts will be left behind, both in search rankings and their use through computational tools. This Catatylst Funded project developed a set of guidelines, and tools to assist organizations in improving their existing OCRed collections, this white paper explores the results of the grant project.