Giovanni Giuffrida - Los Angeles CA, US Eddie Shek - Sherman Oaks CA, US Jihoon Yang - Fairfax VA, US
International Classification:
G06F007/00
US Classification:
707/001000
Abstract:
A spatial knowledge base approach for the automatic extraction of metadata from electronic documents The electronic document is converted to a substantially format invariant data file by an intermediate language conversion element Spatial layout facts are extracted and combined with spatial layout rules from a knowledge engineer in a spatial metadata-reasoning element to provide the metadata The invention is based on mimicking the visual and spatial knowledge that humans make use of when reading a document.