Cross Document Fusion
While Semantex™ provides significant value by extracting content at the document level, it goes even further by consolidating information across documents. This unique capability provides the basis for information discovery, as well as information integration.
CROSS-DOCUMENT ENTITY PROFILES
Cross document entity profile generation results in rich entity summaries that consolidate mentions of an entity across a document collection. This includes entity relationships and any key events associated with the entity. This provides at-a-glance information discovery benefits when processing a large document collection whose contents are not known prior to processing.
There are two capabilities that we address in order to provide this rich source of information.
- The first is name matching, that is, determining that two names are the same regardless of spelling variation or aliases.
- The second, is entity disambiguation which determines whether entity profiles generated from different documents but corresponding to the same name are really the same individual. This answers the question "Is this John Smith the same as the other John Smith"? The unique context provided by Semantex entity profiles along with state-of-the-art machine learning techniques allow us to accurately merge information.
