I want to parse a large number of books < 17th century in search of information about the disappeared people, how to solve the problem?

Question

Parsing

in the case of pdf, I can parse with regular expressions, but "books" will of course be more often scans, and more often some kind of unformatted manuscripts, but in different languages, and moreover, outdated versions of languages. computer vision is likely to be needed, what tool would be chosen for such a task?

Answer 1

Take a look here https://github.com/tesseract-ocr/tesseract

Answer 2

This task is not solved automatically, but only manually.