Artificial intelligence language program

• US 5,587,903 A
• Filed: 06/22/1994
• Issued: 12/24/1996
• Est. Priority Date: 06/22/1994
• Status: Expired due to Fees

• Current Assignee: Thomas W. Yale
• Original Assignee: Thomas W. Yale
• Inventors: Yale, Thomas W.
• Primary Examiner(s): McElheny, Jr., Donald E.
• Assistant Examiner(s): Tkacs, Stephen R.
• Application Number:
  US08/264,071
• Time in Patent Office: 916 Days
• Field of Search: 364/419.01, 364/419.02, 364/419.04, 364/419.08, 395/12, 395/54, 395/55, 395/60, 395/63, 395/67
• US Class Current: 704/9
• CPC Class Codes: G06F 40/55 Rule-based translation

Abstract

A sentence entered by a user is translated into Esperanto. The translated sentence is then parsed and a series of propositions are generated. Each proposition consist of a subject, verb, and an optional object such as an adjective, noun, or prepositional phrase, if present. A series of propositional databases are used to relate the various propositions as to their function within the sentence, thereby mapping the thought of the sentence in a form recognizable by a machine.

Claims

1. A method for translating a sentence entered by a user into machine recognizable thought patterns, said method comprising:

receiving the sentence entered by the user;

translating the user'"'"'s sentence to a universal language;

parsing the translated sentence;

deriving propositions from the parsed translated sentence of the user;

sequentially storing the derived propositions into a knowledge database; and

,

generating a set of peripheral database linkages for the sets of derived propositions representing the user'"'"'s sentence, wherein each peripheral database linkage includes a peripheral database representing a function of speech selectively linked to one or more of the other peripheral databases and the knowledge database.

2. A method as claimed in claim 1, wherein said step of translating includes the steps of resolving any semantic and syntactic ambiguities.

3. A method as claimed in claim 2, further comprising the step of resolving any pronoun ambiguities after said step of parsing the translated sentence.

4. A method as claimed in claim 1, further comprising the steps of storing each noun present in each derived proposition into a noun database listing in conjunction with the identification of the proposition the noun first appears respective to the sequential order of the propositions stored in the knowledge database.

5. A method as claimed in claim 1, further comprising the steps of storing in a logic database the identification of a pair of propositions stored in the knowledge database having a common subject noun.

6. A method as claimed in claim 1, further comprising the steps of storing in an object database the identification of a pair of propositions, wherein a first proposition of the pair of propositions includes an object noun used as a subject noun in a second proposition of the pair of propositions.

7. A method as claimed in claim 1, wherein said step of generating a set of peripheral database linkages include the steps of:

storing in a first number database all database linkages having the same subject, wherein the subject is a number;

storing in a second number database all database linkages having the same object, wherein the object is a number;

storing in an adverb database adverbs found in the knowledge database; and

,

storing in a date database any dates found in the knowledge database;

wherein each listing within any of the peripheral databases includes linking data identifying each database that listing can be found.

8. A system for translating a sentence entered by a user into machine recognizable thought patterns, said method comprising:

means for receiving the sentence entered by the user;

means for translating the user'"'"'s sentence to a universal language;

means for parsing the translated sentence;

means for deriving propositions from the parsed translated sentence of the user;

means for sequentially storing the derived propositions into a knowledge database; and

,

means for generating a set of peripheral database linkages for said sets of derived propositions representing the user'"'"'s sentence, wherein each of said peripheral database linkages include a peripheral database representing a function of speech selectively linked to one or more of the other peripheral databases and the knowledge database.

9. A system as claimed in claim 8, wherein said translation means includes the means for resolving any semantic and syntactic ambiguities.

10. A system as claimed in claim 8, further includes means for resolving any pronoun ambiguities.

11. A system as claimed in claim 8, means for storing each noun present in each derived proposition into a noun database listing in conjunction with the identification of the proposition the noun first appears respective to the sequential order of the propositions stored in the knowledge database.

12. A system as claimed in claim 8, further comprising means for storing in a logic database the identification of a pair of propositions stored in the knowledge database having a common subject noun.

13. A system as claimed in claim 8, further comprising means for storing in an object database the identification of a pair of propositions, wherein a first proposition of the pair of propositions includes an object noun used as a subject noun in a second proposition of the pair of propositions.

14. A system as claimed in claim 8, wherein generating means further includes:

means for storing in a first number database all database linkages having the same subject, wherein the subject is a number;

means for storing in a second number database all database linkages having the same object, wherein the object is a number;

means for storing in an adverb database adverbs found in the knowledge database; and

,

means for storing in a date database any dates found in the knowledge database;

wherein each listing within any of said peripheral databases includes linking data identifying each database that listing can be found.