Method and apparatus for determining the meaning of natural language
First Claim
1. A computer-implemented method for processing natural language, comprising:
- receiving an information input string comprising at least one stimulus;
for each stimulus, generating a referent tridbit comprising (R1) a referent type element defining a stimulus type selected from the group consisting of a property, a thing, and an event, (R2) a referent value element, (R3) a referent scope;
count element, and (R4) a referent attribute element identifying a connection between the referent tridbit and the stimulus;
storing each referent tridbit in computer memory;
generating at least one assert tridbit comprising (A1) a target element that identifies one of the referent tridbits as a target tridbit;
(A2) an assert attribute element; and
(A3) an assert value element, wherein each assert tridbit defines a relationship among the target tridbit, the assert attribute element and the assert value element;
storing the assert tridbits in computer memory; and
validating each assert tridbit by determining that its assert value element is a member of a category defined by its assert attribute element, and that its assert attribute element corresponds to an attribute of the target tridbit.
0 Assignments
0 Petitions
Accused Products
Abstract
A method for processing natural language includes receiving an information input string. Referent tridbits corresponding to stimuli in the information input string are generated. Assert tridbits defining relationships between the referent tridbits are generated. A language processing system including a rules database and a meaning engine. The rules database is operable to store a plurality of syntax rules. The meaning engine is operable to receiving an information input string, generate referent tridbits corresponding to stimuli in the information input string based on the syntax rules, and generate assert tridbits defining relationships between the referent tridbits based on the syntax rules.
391 Citations
14 Claims
-
1. A computer-implemented method for processing natural language, comprising:
-
receiving an information input string comprising at least one stimulus; for each stimulus, generating a referent tridbit comprising (R1) a referent type element defining a stimulus type selected from the group consisting of a property, a thing, and an event, (R2) a referent value element, (R3) a referent scope;
count element, and (R4) a referent attribute element identifying a connection between the referent tridbit and the stimulus;storing each referent tridbit in computer memory; generating at least one assert tridbit comprising (A1) a target element that identifies one of the referent tridbits as a target tridbit;
(A2) an assert attribute element; and
(A3) an assert value element, wherein each assert tridbit defines a relationship among the target tridbit, the assert attribute element and the assert value element;storing the assert tridbits in computer memory; and validating each assert tridbit by determining that its assert value element is a member of a category defined by its assert attribute element, and that its assert attribute element corresponds to an attribute of the target tridbit. - View Dependent Claims (2, 3, 4, 5, 6, 7, 8, 9, 10)
-
-
11. A language processing system, comprising:
-
a rules database operable to store a plurality of syntax rules; and a meaning engine operable to receive an information input string, to generate a referent tridbit comprising (R1) a referent type element defining a stimulus type selected from the group consisting of a property, a thing, and an event, (R2) a referent value element, (R3) a referent scope;
count element, and (R4) a referent attribute element identifying a connection between the referent tridbit and the stimulus based on the syntax rules and to generate at least one assert tridbit comprising (A1) a target element that identifies the referent tridbit as a target tridbit;
(A2) an attribute element defining a referent of general scope; and
(A3) an assert value element defining a category represented by the referent of the attribute element, each assert tridbit defines a relationship among the target tridbit, the assert attribute element and the assert value based on the syntax rules.
-
-
12. A knowledge processing system, comprising:
-
a knowledgebase operable to store referent tridbits corresponding to stimuli and assert tridbits defining relationships between the referent tridbits; and a meaning engine operable to access the knowledgebase and perform a reasoning process using at least one referent tridbit comprising (R1) a referent type element defining a stimulus type selected from the group consisting of a property, a thing, and an event, (R2) a referent value element, (R3) a referent scope;
count element, and (R4) a referent attribute element identifying a connection between the referent tridbit and the stimulus based on the syntax rules, and to generate at least one assert tridbit comprising (A1) a target element that identifies the referent tridbit as a target tridbit;
(A2) an attribute element defining a referent of general scope; and
(A3) an assert value element defining a category represented by the referent of the attribute element, each assert tridbit defines a relationship among the target tridbit, the assert attribute element and the assert value.
-
-
13. A system for processing natural language, comprising:
-
means for receiving an information input string; means for generating a referent tridbit comprising (R1) a referent type element defining a stimulus type selected from the group consisting of a property, a thing, and an event, (R2) a referent value element, (R3) a referent scope;
count element, and (R4) a referent attribute element identifying a connection between the referent tridbit and the stimulus;means for generating assert tridbits comprising (A1) a target element that identifies the referent tridbits as a target tridbit;
(A2) an assert attribute element; and
(A3) an assert value element, wherein each assert tridbit defines a relationship among the target tridbit, the assert attribute element and a the assert value element;and means for validating each assert tridbit by determining that its assert value element is a member of a category defined by its assert attribute element, and that its assert attribute element corresponds to an attribute of the target tridbit.
-
-
14. A program storage device encoded with instructions that, when executed by a processing device, perform a method for processing natural language, the method comprising:
-
receiving an information input string comprising at least one stimulus; for each stimulus, generating a referent tridbit comprising (R1) a referent type element defining a stimulus type selected from the group consisting of a property, a thing, and an event, (R2) a referent value element, (R3) a referent scope;
count element, and (R4) a referent attribute element identifying a connection between the referent tridbit and the stimulus;generating at least one assert tridbit comprising (A1) a target element that identifies one of the referent tridbits as a target tridbit;
(A2) an assert attribute element; and
(A3) an assert value element, wherein each assert tridbit defines a relationship among the target tridbit, the assert attribute element and the assert value element; andvalidating each assert tridbit by determining that its assert value element is a member of a category defined by its assert attribute element, and that its assert attribute element corresponds to an attribute of the target tridbit.
-
Specification