System and method of grouping and extracting information from data corpora

• US 9,588,963 B2
• Filed: 04/03/2015
• Issued: 03/07/2017
• Est. Priority Date: 03/18/2009
• Status: Active Grant

• Current Assignee: Entigenlogic LLC
• Original Assignee: Iqintell, Inc.
• Inventors: Williams, Walter Edward, Williams, Frank John
• Primary Examiner(s): GUERRA-ERAZO, EDGAR X
• Application Number:
  US14/678,294
• Publication Number:
  US 20150363384A1
• Time in Patent Office: 704 Days
• Field of Search: 704/9, 704/10, 704/270, 704/270.1, 704/275
• US Class Current: 1/1
• CPC Class Codes: G06F 40/30 Semantic analysis

Abstract

A system for annotating words of a data corpus based upon their particular concept and their corresponding grammatical sense with Conceptual Numerical Identifiers (CNIs) from a Conceptual Dictionary, pairing the words based on conceptual inter-relating network (CIRN) rules, and determining if a selected plurality of paired words are grammatically, syntactically, and linguistically correct by matching CNIs from each pair of words.

Claims

1. A method of extracting information from data corpora, the method comprising:

identifying a data corpus accessible by a computer comprising grammatical sentences made up of a plurality of word elements;

assigning conceptual numerical identifiers to the plurality of word elements by a processor, where the conceptual numerical identifiers each has a word portion and a meaning portion;

grouping the conceptual numerical identifier based upon predetermined rules that group the conceptual numerical identifiers to form sets with conceptual sets logic, where the groupings of the conceptual numerical identifiers are stored in memory;

applying additional rules to the groupings of the conceptual numerical identifiers by the processor that result in indexed groups of conceptual numerical identifier; and

storing the indexed groupings of conceptual numerical identifier in a database, where the indexed groupings of conceptual numerical identifier may be queried with a grammatical sentence.

2. The method of claim 1, further includes accessing a conceptual dictionary to identify the conceptual numerical identifiers for word elements in the plurality of word elements.

3. The method of claim 1, where assigning conceptual numerical identifiers includes generating a user prompt to identify a meaning of an unidentified word element that is not in the conceptual dictionary.

4. The method of claim 1, where the predetermined rules include grammar rules.

5. The method of claim 4 further includes, inserting the unidentified word element into the conceptual database along with the conceptual numerical identifier associated with the word element in response to information entered in response to a user prompt.

6. The method of claim 1, where assigning conceptual numerical identifiers includes identifying ambiguity between word meanings and prompting a user to select a meaning in order for a selection to be made between conceptual numerical identifiers for word elements that have multiple definitions.

7. The method of claim 1, includes converting the grammatical sentence query into groups of conceptual numerical identifiers that are used to query the indexed groupings of conceptual numerical identifiers stored in the database.

8. The method of claim 7, further includes generating a result to the query, and

converting the result back into a grammatically correct answer from the conceptual numerical identifiers.

15. The method of claim 1, further includes generating a result to the query, and

converting the result back into a grammatical correct answer from the conceptual numerical identifiers.

9. A method of extracting information from data corpora, the method comprising:

converting a query made up of a grammatical string into a plurality of word elements;

assigning conceptual numerical identifiers to word elements in the plurality of word elements, such that the conceptual numerical identifiers identifies a specific word meaning to the word elements, where the conceptual numerical identifiers each has a word portion and a meaning portion;

grouping the conceptual numerical identifiers based upon predetermined rules that group the conceptual numerical identifiers to form sets with conceptual sets logic, where the groupings of the conceptual numerical identifiers are stored in memory;

applying additional rules to the groupings of the conceptual numerical identifiers by a processor that result in indexed grouping of numerical identifier; and

accessing a database with the indexed groupings of numerical identifiers, where the database contains groupings of conceptual numerical identifiers associated with the data corpora in order to identify matches between a portion of the indexed groupings of numerical identifiers with the groupings of conceptual numerical identifiers associated with the data corpora.

10. The method of claim 9, where the predetermined rules include grammar rules.

11. The method of claim 9, further includes accessing a conceptual dictionary to identify the conceptual numerical identifiers for word elements in the plurality of word elements.

12. The method of claim 11, where grouping conceptual numerical identifiers includes generating a user prompt to identify a meaning of an unidentified word element that is not in the conceptual dictionary.

13. The method of claim 12 further includes, inserting the unidentified word element into the conceptual database along with the numerical identifier associated with the word element in response to information entered in response to the user prompt.

14. The method of claim 9, where assigning conceptual numerical identifiers includes identifying ambiguity between word meanings and prompting a user to select a meaning in order for a selection to be made between numerical identifiers for word elements that have multiple definitions.

16. A method of extracting information from data corpora, the method comprising:

identifying a data corpus accessible by a computer comprising grammatical sentences made up of a plurality of word elements;

assigning conceptual numerical identifiers to the plurality of word elements by a processor, where the conceptual numerical identifiers each has a word portion and a meaning portion;

grouping the conceptual numerical identifiers into pairings of conceptual numerical identifiers based upon predetermined rules that group the conceptual numerical identifiers to form sets with conceptual sets logic, where the pairings are stored in memory;

applying additional rules to the pairings of the conceptual numerical identifiers by the processor that result in indexed pairs of conceptual numerical identifiers; and

storing the indexed pairings of conceptual numerical identifiers in a database, where the indexed pairings of conceptual numerical identifier may be queried with a grammatical sentence.

17. The method of claim 16, further includes accessing a conceptual dictionary to identify the conceptual numerical identifiers for word elements in the plurality of word elements.

18. The method of claim 16, where assigning conceptual numerical identifiers includes generating a user prompt to identify a meaning of an unidentified word element that is not in the conceptual dictionary.

19. The method of claim 16, where the predetermined rules include grammar rules.

20. The method of claim 19 further includes, inserting the unidentified word element into the conceptual database along with the conceptual numerical identifiers associated with the word element in response to information entered in response to a user prompt.

21. The method of claim 16, where assigning conceptual numerical identifiers includes identifying ambiguity between word meanings and prompting a user to select a meaning in order for a selection to be made between conceptual numerical identifiers for word elements that have multiple definitions.

22. The method of claim 16, includes converting the grammatical sentence query into pairings of conceptual numerical identifiers that are used to query the indexed pairings of conceptual numerical identifiers stored in the database.

23. The method of claim 22, further includes generating a result of conceptual numerical identifiers to the query, and

converting the result of conceptual numerical identifiers back into a grammatically correct answer.

24. A non-transitory machine-readable medium with machine readable instructions, that when executed result in a method of extracting information from data corpora, the instructions for the steps comprising:

identifying a data corpus accessible by a computer comprising grammatical sentences made up of a plurality of word elements;

assigning conceptual numerical identifiers to the plurality of word elements by a processor, where the conceptual numerical identifiers each has a word portion and a meaning portion;

grouping of the conceptual numerical identifiers based upon predetermined rules that group the conceptual numerical identifiers to form sets with conceptual sets logic, where the groupings of the conceptual numerical identifiers are stored in memory;

applying additional rules to the groupings the conceptual numerical identifiers by the processor that result in indexed groups of conceptual numerical identifiers; and

storing the indexed groupings of conceptual numerical identifiers in a database, where the indexed groupings of conceptual numerical identifiers may be queried with a grammatical sentence.

25. The non-transitory machine-readable medium with machine readable instructions of claim 24, further includes instructions for accessing a conceptual dictionary to identify the conceptual numerical identifiers for word elements in the plurality of word elements.

26. The non-transitory machine-readable medium with machine readable instructions of claim 25, where assigning conceptual numerical identifiers includes generating a user prompt to identify a meaning of an unidentified word element that is not in the conceptual dictionary.

27. The non-transitory machine-readable medium with machine readable instructions of claim 24, where the predetermined rules include grammar rules.

28. The non-transitory machine-readable medium with machine readable instructions of claim 27 further includes instructions for inserting the unidentified word element into the conceptual dictionary along with the conceptual numerical identifier associated with the word element in response to information entered in response to a user prompt.

29. The non-transitory machine-readable medium with machine readable instructions of claim 24, where assigning conceptual numerical identifiers includes identifying ambiguity between word meanings and prompting a user to select a meaning in order for a selection to be made between conceptual numerical identifiers for word elements that have multiple definitions.

30. The non-transitory machine-readable medium with machine readable instructions of claim 25, includes instructions for converting the grammatical sentence query into groups of conceptual numerical identifiers that are used to query the indexed groupings of conceptual numerical identifiers stored in the database.

31. The non-transient machine readable media with machine readable instructions of claim 30, further includes:

instructions for generating a result to the query, and

converting the result back into a grammatically correct answer from the conceptual numerical identifiers.