In solutions where text archives are digitized to become searchable in Watson Discovery, Smart Document Understanding is a promising new feature. However, it does not by side effect produce a searchable (text-based) PDF when OCRing a scanned PDF. This is a feature of IBM's BACA system. It allows for user to open, after finding a relevant document via Discovery, what looks like the original scanned PDF but as a searchable/text-based PDF where user can now apply control-F to search where the term or phrase of interest occurs, and copy text passages into clipboard etc.
Potential benefit is not estimated, but the benefit is clear for any Discovery use case when searching large enterprise collections that have been OCRd.
Why is it useful?
|Who would benefit from this IDEA?||As an analyst, I have searched my company's digitized archives (or previous projects, history, contracts) and now am able to search within the original document the larger context of the passages matching my search results.|
How should it work?