Method and system for machine understanding, knowledge, and conversation

• US 8,117,022 B2
• Filed: 12/07/2007
• Issued: 02/14/2012
• Est. Priority Date: 12/07/2006
• Status: Active Grant

• Current Assignee: Sheldon O. Linker
• Original Assignee: Sheldon O. Linker
• Inventors: Linker, Sheldon O.
• Primary Examiner(s): Godbold, Douglas
• Application Number:
  US11/952,796
• Publication Number:
  US 20080140387A1
• Time in Patent Office: 1,530 Days
• Field of Search: 704/1, 704/9, 704/10, 704/46
• US Class Current: 704/9
• CPC Class Codes: G06F 40/40 Processing or translation of natural language natural language analysis G06F40/20; semantic analysis G06F40/30

Abstract

A system for storing knowledge in a speakable, formal class of languages and allowing querying and retrieving of that knowledge is provided. The system further allows the users to input data, instructions, questions, and queries without structure or rules over and above the actual grammar of the class of languages. The system responds in words, phrases or complete sentences as applicable, and its output is bound by the rules of logic, rather than simple correlation or likely meaning.

Claims

1. A method operational in a processing device, comprising:

inputting data into the processing device in a formal, naturalistic, speakable, hybrid language, wherein the formal, naturalistic, speakable, hybrid language is characterized by having a completely defined set of non-ambiguous grammar rules and excludes natural languages;

parsing the input data at the processing device into an internal format in such a way that a memory of an original grammar structure present in the input data is retained;

comparing the parsed input data with previously stored data in a data/knowledge base according to non-statistical, logic operations to prove, disprove or determine if the meaning of the parsed input data is plausible;

storing results of the evaluation in the data/knowledge base, wherein the input data is obtained from a plurality of different users and the data/knowledge base stores knowledge while recognizing, maintaining, and storing distinct realities for different users; and

in reply to a question, generating one of an auditory or visual response at the processing device in the formal, naturalistic, speakable, hybrid language based on accumulated knowledge of the data/knowledge base,

wherein for the same question, a first auditory or visual response for a first user is contradictory to a second auditory or visual response for a second user if the first and second users have different realities.

2. The method of claim 1, wherein the formal, naturalistic, speakable, hybrid language is understood by speakers of a corresponding natural language having ambiguous grammar rules.

3. The method of claim 1, wherein the formal, naturalistic, speakable, hybrid language is based on a grammar structure for a formal language and vocabulary of a natural language.

4. The method of claim 1, wherein the formal, naturalistic, speakable, hybrid language is characterized by at least one of a Loglan-compatible or Lojban-compatible language structure.

5. The method of claim 1, further comprising

comparing a first utterance to a second utterance in an open-loop, where the first utterance has a higher level of specificity than the second utterance, to determine if the second utterance is superfluous.

6. The method of claim 1, wherein the formal, naturalistic, speakable, hybrid language includes a grammar structure defined by:

a first characteristic to form predications and statements;

a second characteristic to form questions and queries;

a third characteristic to apply modifiers to portions of knowledge in the data/knowledge base to formulate the response; and

a fourth characteristic ability to apply a broad category of tenses to portions of the text.

7. The method of claim 1, further comprising retaining the input data if the input data is a statement which provides new information to the data/knowledge base and adding the statement to the data/knowledge base.

8. The method of claim 7, wherein the statement is added to the data/knowledge base based upon rules of veracity, conjunctions, implications, and disjunctions, where veracity is achieved based on a veracity level associated with a user providing the statement.

9. The method of claim 1, wherein all forms of knowledge are stored together in the data/knowledge base as predications.

10. The method of claim 1, wherein generating the response in the formal, naturalistic, speakable, hybrid language includes utilizing adjectival phrases to qualify the knowledge.

11. The method of claim 1, wherein generating the response in the formal, naturalistic, speakable, hybrid language, time-sensitive context of knowledge in the data/knowledge base is considered in formulating the response.

12. The method of claim 1, wherein such distinct realities for different users includes contradictory conclusions for different users.

13. The method of claim 1, wherein different veracity levels are associated with the different users, such veracity levels are utilized in generating the response from inconsistent knowledge stored in the data/knowledge base.

14. The method of claim 1, wherein the response is in full sentence format.

15. The method of claim 1, wherein the data/knowledge base concurrently stores contradictory knowledge for different users.

16. A device for processing a language selected by a user, comprising:

an input interface configured to receive input data in the form of utterances in a formal, naturalistic, speakable, hybrid language, wherein the formal, naturalistic, speakable, hybrid language is characterized by having a completely defined set of non-ambiguous grammar rules and excludes natural languages;

a processing circuit adapted to implement

a parsing module adapted to parse the input data from the input module into an internal format;

a prover module adapted to receive the formatted input data from the parse module and comparing the parsed input data with previously stored data in a data/knowledge base according to non-statistical, logic operations to prove, disprove or determine if the parsed input data is plausible; and

a memory device configured to store results of the prover module in the data/knowledge base, wherein the input data is obtained from a plurality of different users and the data/knowledge base stores knowledge while recognizing, maintaining, and storing distinct realities for different users;

wherein in reply to a question, the processing circuit further generates one of an auditory or visual response at the processing device in the formal, naturalistic, speakable, hybrid language based on accumulated knowledge of the data/knowledge base,

wherein for the same question, a first auditory or visual response for a first user is contradictory to a second auditory or visual response for a second user if the first and second users have different realities.

17. The device of claim 16, wherein the prover module is configured use the non-statistical, logic operations to prove the statements, disprove the statements or show that the meaning of the statements are plausible and saves in the data/knowledge base.

18. A method operational in a processing device, comprising:

inputting data into the processing device in a formal, naturalistic, speakable, hybrid language, wherein the formal, naturalistic, speakable, hybrid language is characterized by having a completely defined set of non-ambiguous grammar rules and excludes natural languages;

parsing the input data at the processing device into an internal format in such a way that a memory of an original grammar structure present in the input data is retained;

comparing the parsed input data with previously stored data in a data/knowledge base according to non-statistical, logic operations to prove, disprove or determine if the meaning of the parsed input data is plausible;

storing results of the evaluation in the data/knowledge base, wherein the input data is obtained from a plurality of different users and the data/knowledge base stores knowledge while recognizing, maintaining, and storing distinct realities for different users; and

in reply to a question, generating one of an auditory or visual response at the processing device in the formal, naturalistic, speakable, hybrid language based on accumulated knowledge of the data/knowledge base,

wherein the auditory or visual response to a first user includes contradictory realities from different users if first user does not have a stored reality with which to provide the response.