AUGMENTED EXPLORATION FOR BIG DATA AND BEYOND

摘要

A computer is to obtain specification concept graphs of nodes spec1, spec2, . . . , specm including concept nodes and relation nodes according to at least one of a plurality of digitized data from a plurality of computerized data sources forming a first set of evidences U and obtain concept graphs of nodes cα1, cα2, . . . , cαn including concept nodes and relation nodes for corresponding obtained plurality of information and knowledge (IKs) α1,α2, . . . , αn forming a second set of evidences U. A subset of concept graphs of nodes is selected from cα1, cα2, . . . , cαn according to a computable measure of consistency, inconsistency and/or priority threshold between cαj in cα1, cα2, . . . , cαn can to specification concept graph speck in spec1, spec2, . . . , spec m. Knowledge fragments are generated for corresponding subset of concept graphs cαi1, cαi2, . . . , cαih include augmenting information objects by creating or adding into at least one knowledge-base (KB), new objects in form ω=E→A from the concept; fragments, including a computed validity (v) and a plausibility (p) for a combination of relationship constraints k for the concept fragments and obtained propositions k for the fragment concepts.