RSS

One day the paperless office will prevail, but currently organizations still create and receive “metres” of printed documents daily. Their challenge is to digitize these documents into an easily retrievable location while keeping the manual intervention and costs at minimum.

Pingar and Fuji Xerox have jointly created a solution to this problem: Users can now scan bulks of paper documents straight into the document management system. All necessary metadata is generated also in bulk and on the fly. The result is a well organized library of documents sorted by categories, each with a set of specific keywords and entities mentioned in text which aids searching.

The original way to digitize a document is to scan one page at a time into multiple images that are then combined into a single PDF document. The optical character recognition (OCR) technology, continually improved since its development in late 70s, recognizes the outlines of letters and numbers and converts them into digital text. This way, the documents require less storage space, and can be edited and searched. But where are the documents saved to? How can we retrieve them later efficiently?

The Apeos SmartConnect application by Fuji Xerox provides a touch-screen interface which helps storing scanned documents into enterprise intranet portals. The user separates multiple documents using divider pages, and SmartConnect generates multiple PDF documents, which are then added into systems like Microsoft SharePoint. The Pingar API extends this application with one extra step: The text extracted through OCR is not just added into the PDF, but also sent to the Pingar’s semantic engine. A series of Pingar entity extraction tools are fired at that text and return back key categories, terms, keywords and entities that describe the document’s content. SmartConnect stores these as the document’s metadata in the associated library. When working with this library, users don’t need to open the document, in order to see whether it’s relevant. When searching in this library, users can refine results by categories and entities.

 Watch this video to see how easy and effective this process works in real life:

Comments: