Combinatorial computational technique for transformation phrase text-phrase meaning

• US 6,401,061 B1
• Filed: 06/28/1999
• Issued: 06/04/2002
• Est. Priority Date: 05/13/1999
• Status: Expired due to Fees

• Current Assignee: Microsoft Technology Licensing LLC (Microsoft Corporation)
• Original Assignee: Yuri L. Zieman
• Inventors: Zieman, Yuri L.
• Primary Examiner(s): EDOUARD, PATRICK NESTOR
• Application Number:
  US09/342,029
• Time in Patent Office: 1,072 Days
• Field of Search: 704/1, 704/9, 704/10, 707/500, 707/530, 707/532, 707/1, 707/2, 707/103, 707/104
• US Class Current: 704/10
• CPC Class Codes: G06F 40/268 Morphological analysis
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

A combinatorial system for extracting major meaning components of a phrase or sentence text in natural language and vice versa. This system allows more intelligent processing and retrieval of textual information in computers. More specifically, the system is divided into three parts. The first is based on a specially defined set of universal primary or atomic abstract concepts, Semantic Factors and rules and methods for assembling and modifying the Semantic Factors. The second is based on a specially defined set of morpheme-type linguistic elements of a specific language, referred to herein as S-Morphs, that are formed into a dictionary. The third part relies upon algorithms and rules for using the two parts described above to translate text into its conceptual description or meanings and/or point from a meaning to relevant objects.

Claims

1. Apparatus for relating meanings and text, said text associated with a given language, said apparatus including a computing system, comprising:

means for forming a set of concepts wherein said concepts relate to meanings for any language,

means for storing said concepts within the computing system,

means for forming morpheme-type linguistic elements specific to said given language,

means for storing the linguistic elements in the computing system,

dictionary means for associating said concepts with the linguistic elements,

means for splitting said text into the linguistic elements, and

means for looking up the linguistic elements in said dictionary and retrieving the related concepts, thereby relating said meanings to said text.

2. The apparatus as defined in claim 1 wherein said dictionary means for associating the concepts with linguistic elements includes multiple alternative concepts for the same linguistic element.

3. The apparatus as defined in claim 1 wherein said concepts comprise independent concepts of human understanding.

4. The apparatus as defined in claim 1 further comprising modifiers for said concepts, wherein said modifiers add quantitative and comparative information about said concepts, and wherein said modifiers are retained with said related concepts.

5. The apparatus as defined in claim 1 wherein said linguistic elements comprise:

first type linguistic elements identified with the beginning of a word,

second type linguistic elements identified with any position within a word,

third type linguistic elements identified with the end of a word, and

fourth type linguistic elements identified with a separate word.

6. The apparatus as defined in claim 5 wherein said linguistic elements comprise fifth type linguistic elements identified with abbreviations.

7. The apparatus as defined in claims 1 wherein said text is of any human language.

8. The apparatus as defined in claim 1 wherein said means for splitting said text into linguistic elements comprises means for processing said text in both a left-to-right and a right-to-left direction.

9. The apparatus as defined in claim 1 wherein when said linguistic element is not found in said dictionary, said process omits that linguistic element and continues processing.

10. The apparatus as defined in claim 1 further comprising:

means for forming groups of said concepts;

means for defining objects with said groups; and

means for matching the concepts relating to meanings from said text with said groups, whereby objects are determined to exist or not exist in said text.

11. The apparatus as defined in claim 1 further comprising:

means for associating a group of concepts with a subject object, and

means for matching said group of concepts associated with said subject object with said concepts associated with said linguistic elements, wherein, when a match is found, said subject object is determined to be within the meaning of said text, and when a match is not found, said subject object is determined not to be within the meaning of said text.

12. The apparatus as defined in claim 11 wherein said means for matching comprises:

means for determining when alternative meanings are present in said dictionary, and

means for choosing one of the alternative meanings, thereby providing no ambiguity associated with multiple meanings of said text.

13. The apparatus as defined in claim 11 further comprising a threshold for determining when a match is found, said threshold defines a given number of common concepts found that are associated with the linguistic elements and with said subject object.

14. A computer implemented process for relating meanings and text, said text associated with a given language, comprising the steps of:

forming a set of concepts wherein said concepts relate to meanings of any language,

storing said concepts within a computer system,

forming a set of morpheme-type linguistic elements specific to said given language,

storing the linguistic elements within the computer system,

forming a dictionary relating the linguistic elements with said concepts,

splitting said text into linguistic elements, and looking up the linguistic elements in said dictionary and retrieving the related concepts, thereby relating said meanings to said text.

15. The computer implemented process as defined in claim 14 wherein the step of forming a dictionary includes relating multiple alternative concepts to the same linguistic element.

16. The computer implemented process as defined in claim 14 further comprising the steps of:

forming groups of said concepts;

defining objects with said groups; and

matching the concepts relating to meanings from said text with said groups, whereby objects are determined to exist or not exist in said text.

17. The computer implemented process as defined in claim 14 further comprising the step of forming said concepts as independent concepts of human understanding.

18. The computer implemented process as defined in claim 14 further comprising the step of defining modifiers for said concepts, wherein said modifiers add quantitative and comparative information about said concepts, and wherein said modifiers are retained with said related concepts.

19. The computer implemented process as defined in claim 14 further comprising the steps of:

identifying first linguistic elements with the beginning of a word,

identifying second linguistic elements with any position within a word,

identifying third linguistic elements with the end of a word, and

identifying fourth linguistic elements with separate words.

20. The computer implemented process as defined in claim 19 further comprising the step of identifying fifth linguistic elements identified with abbreviations.

21. The computer implemented process as defined in claim 14 further comprising the step of determining that the text is of any human language.

22. The computer implemented process as defined in claim 14 wherein said step of splitting said text into linguistic elements comprises the step of processing said text in both a left to right and a right to left direction.

23. The computer implemented process as defined in claim 14 wherein when, in the step of looking up the linguistic element, a linguistic element is not found in said dictionary means for associating, said process omits that linguistic element and continues processing.

24. The computer implemented process as defined in claim 14 further comprising the steps of:

associating a group of concepts with a subject object,

matching said group of concepts associated with said subject object with said concepts associated with said linguistic elements, wherein, when a match is found, said subject object is determined to be within the meaning of said text, and when a match is not found, said subject object is determined not to be within the meaning of said text.

25. The computer implemented process as defined in claim 24 further comprising the steps of:

determining when alternative meanings are present in said dictionary, and

choosing one of the alternative meanings, thereby providing no ambiguity associated with multiple meanings of said text.

26. The computer implemented process as defined in claim 24 further comprising the step of defining a threshold, wherein a match is found when said threshold is met, wherein said threshold defines a given number of common concepts found that are associated with the linguistic elements and with said subject object.

27. A computer-readable medium bearing a program code embodied therein for relating meanings and text, said text associated with a given language, and including:

a first computer readable file for forming a set of concepts wherein said concepts relate to meanings of any language,

a second computer readable file for storing said concepts within a computer system,

a third computer readable file for forming a set of morpheme-type linguistic elements for said given language,

a fourth computer readable file for storing said linguistic elements,

a fifth computer readable file defining a dictionary for associating said concepts with linguistic elements,

a sixth computer readable file—

program for splitting said text into the linguistic elements,

a seventh computer readable file defining subject objects with groups of said concepts, said groups defining a semantic index, and

an eighth computer readable file—

program for looking up said groups of concepts in said semantic index and retrieving the related objects, thereby relating said meanings to said objects.