Knowledge base with clustered data objects
AssigneeFair Isaac Corporation
InventorsJohn Byrnes, Richard Rohwer
Life of this patent
Estimating the term…
Abstract
A knowledge base is first characterized by an association-grounded semantics collapsed language. In response to the receipt of a query of the knowledge base, the collapsed language is used to determine whether there is an indication that the knowledge base contains knowledge requested in the query. Thereafter, the collapsed language can be used to carry out a full search for the knowledge much more efficiently than would otherwise be possible. Related methods, apparatus, and articles are also described.
Claims20
An article comprising a tangibly embodied machine-readable medium operable to cause one or more machines to result in operations comprising:
- receiving a query of a knowledge base, the knowledge base comprising a plurality of logical symbols compressed into a plurality of disjoint bins;
- reasoning over the plurality of bins to determine whether the bins indicate that the knowledge base contains knowledge requested in the query;
- and
- providing an indication characterizing a likelihood whether the knowledge base contains the knowledge requested in the query.
An article comp rising a tangibly embodied machine-readable medium operable to cause one or more machines to result in operations comprising:
- receiving a query of a knowledge base characterized by an association-grounded semantics collapsed language;
- determining whether there is an indication that the knowledge base contains knowledge requested in the query;
- and
- providing an indication characterizing a likelihood whether the knowledge base contains the knowledge requested in the query.
An article for merging two or more knowledge bases comprising a tangibly embodied machine-readable medium operable to cause one or more machines to result in operations comprising:
- applying association-grounded semantics to collapse a language of each of the knowledge bases to obtain a single class of predicates for each language;
- and
- mapping the classes of predicates for the knowledge bases into a common language to generate a merged knowledge base.
A method comprising:
- receiving a knowledge base including a plurality of logical symbols arranged in formulas;
- assigning a probability distribution to the symbols;
- placing symbols into bins based on a similarity of the assigned probability distributions;
- and
- providing access to a compressed knowledge base in which the bins are used as a reduced set of symbols to enable one or more queries to be performed against the knowledge base.
Dependencies are inferred from the claim text.