Traditionally, intelligent document processing relied on OCR and rule-based extraction. These methods could tell organizations what was on the page but could not tell them where on the page a ...