Large scale probabilistic ontology reasoning

Techniques for computing a solution to a query formulated against a knowledge base (KB) are provided. The techniques include receiving a query formulated against a knowledge base, wherein the knowledge base comprises a set of one or more axioms, wherein each axiom is annotated with a specific probab...

Full description

Saved in:
Bibliographic Details
Main Authors FOKOUE-NKOUTCHE ACHILLE B, KALYANPUR ADITYA, SCHONBERG EDITH G, SRINIVAS KAVITHA
Format Patent
LanguageEnglish
Published 07.06.2016
Subjects
CALCULATING
COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS
COMPUTING
COUNTING
PHYSICS
Online AccessGet full text

Cover

More Information
Summary:Techniques for computing a solution to a query formulated against a knowledge base (KB) are provided. The techniques include receiving a query formulated against a knowledge base, wherein the knowledge base comprises a set of one or more axioms, wherein each axiom is annotated with a specific probability value indicating a degree of certainty assigned thereto, ignoring each probability value of the one or more axioms and computing a solution to the query, computing each of one or more justifications for the query solution, wherein computing each of one or more justifications for the query solution comprises determining a minimal set of one or more axioms in the knowledge base that entail the query solution, and using each probability value of the one or more axioms in each justification to compute a net probability of an inferred query solution.
Bibliography:Application Number: US20090574237