Processing Paper Documents with WISDOM

Donato Malerba, Floriana Esposito, Giovanni Semeraro, and Luca De Filippis
Dipartimento di Informatica - Universita' degli Studi di Bari
via Orabona, 4 - 70126 Bari - Italy
{malerba | esposito | semeraro}

Abstract: WISDOM is a paper-computer interface that can transform printed information into a symbolic representation. This is done into four distinct steps: Document analysis, document classification, document understanding, and text recognition with an OCR. Machine learning tools and techniques are used in the first three steps to easily customize the interface on the exigencies of different users.