Data analysis based on data linking elements

• US 9,053,145 B2
• Filed: 03/06/2013
• Issued: 06/09/2015
• Est. Priority Date: 09/13/2006
• Status: Active Grant

• Current Assignee: InterSystems Corporation
• Original Assignee: InterSystems Corporation
• Inventors: Brands, Michael Rik Frans, Van Hyfte, Dirk Medard Helena
• Primary Examiner(s): Beausoliel, Jr., Robert
• Assistant Examiner(s): Rayyan, Susan F
• Application Number:
  US13/786,533
• Publication Number:
  US 20130311499A1
• Time in Patent Office: 825 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06F 16/24 Querying
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: Y10S 707/913 Multimedia

Abstract

A computer controlled method for automatically segmenting an ensemble of data. The method starts by acquiring an ensemble of data and data is segmented by identifying a first subset of sequences of Linking Data Elements based on a repository of Linking Data Elements. A second subset of sequences of Information Carrying Data Elements is identified, wherein the sequences are linked by the Linking Data Elements. The subsets are provided in a structured format.

Claims

1. At least one non-transitory computer-readable storage medium comprising computer-executable instructions, that when executed by a processor of a computer system, perform a method for analyzing unstructured data, wherein said unstructured data comprises a plurality of data elements, and wherein the unstructured data is at least one of text data, linguistic data, image data, video data, sound data, control data, measurement data, olfactive data and tactile data, the method comprising acts of:

identifying a first subset of the plurality of data elements as Linking Data Elements (LDEs), wherein each of the LDEs is at least one of text data, linguistic data, image data, video data, sound data, control data, measurement data, olfactive data and tactile data, wherein identifying a first subset comprises;

retrieving, from a repository, a predetermined set of stored LDEs; and

comparing the unstructured data to the predetermined set of stored LDEs;

identifying a second subset of the plurality of data elements as Information Carrying Data Elements (ICDEs), wherein the second subset is comprised of every data element not identified as an LDE; and

representing the unstructured data in a structured format based on the identification of LDEs and ICDEs.

2. The at least one non-transitory computer-readable storage medium of claim 1, wherein the method further comprises acts of:

identifying a plurality of LDE sequences within the first subset, each LDE sequence of the plurality of LDE sequences comprising at least one LDE; and

identifying a plurality of ICDE sequences within the second subset, each ICDE sequence of the plurality of ICDE sequences comprising at least one ICDE.

3. The at least one non-transitory computer-readable storage medium of claim 2, wherein the act of representing the unstructured data comprises:

ranking the plurality of LDE sequences and/or the plurality of ICDE sequences based on a property of each of the sequences; and

representing the unstructured data in a structured format based on the act of ranking.

4. The at least one non-transitory computer-readable storage medium of claim 3, wherein the property comprises at least one of:

a length associated with each of the sequences;

an order of occurrence of each of the sequences in the unstructured data;

a frequency of occurrence of each of the sequences in the unstructured data; and

one or more links to other sequences.

5. The at least one non-transitory computer-readable storage medium of claim 2, wherein the method further comprises acts of:

identifying one or more subsequences within the plurality of LDE sequences and/or the plurality of ICDE sequences, wherein the one or more subsequences occur in at least two different sequences of the plurality of LDE sequences and/or the plurality of ICDE sequences.

6. The at least one non-transitory computer-readable storage medium of claim 5, wherein the act of representing the unstructured data comprises:

ranking the plurality of LDE sequences and/or the plurality of ICDE sequences based on a property of the one or more subsequences; and

representing the unstructured data in a structured format based on the act of ranking.

7. The at least one non-transitory computer-readable storage medium of claim 6, wherein the property comprises at least one of:

a length associated with the one or more subsequences;

an order of occurrence of the one or more subsequences in the unstructured data;

a frequency of occurrence of the one or more subsequences in the unstructured data; and

one or more links to other subsequences.

8. The at least one non-transitory computer-readable storage medium of claim 5, wherein the act of representing the unstructured data comprises:

ranking the plurality of LDE sequences and/or the plurality of ICDE sequences based on co-occurring subsequences; and

representing the unstructured data in a structured format based on the act of ranking.

9. The at least one non-transitory computer-readable storage medium of claim 1, wherein the method further comprises acts of:

retrieving, from the repository, predefined information associated with at least one LDE of the predetermined set of stored LDEs and;

ranking the LDE sequences in the first subset based on the predefined information; and

wherein the act of representing comprises representing the unstructured data in a structured format based on the act of ranking.

10. The at least one non-transitory computer-readable storage medium of claim 9, wherein the predefined information comprises a semantic denotation of the at least one LDE.

11. The at least one non-transitory computer-readable storage medium of claim 9, wherein the predefined information comprises a qualification in a qualification system of the at least one LDE.

12. The at least one non-transitory computer-readable storage medium of claim 11, wherein the qualification comprises at least one of:

an indication of mood, an indication of judgment on a positive/negative scale, and a number.

19. The computer controlled system of claim 6, wherein the property comprises at least one of:

a length associated with the one or more subsequences;

an order of occurrence of the one or more subsequences in the unstructured data;

a frequency of occurrence of the one or more subsequences in the unstructured data; and

one or more links to other subsequences.

13. A computer controlled system comprising:

at least one storage device configured to store a predetermined set of stored Linking Data Elements (LDEs); and

at least one processor configured to perform a method for analyzing unstructured data, the method comprising acts of;

acquiring said unstructured data comprising a plurality of data elements, wherein the unstructured data is at least one of text data, linguistic data, image data, video data, sound data, control data, measurement data, olfactive data and tactile data;

identifying a first subset of the plurality of data elements as Linking Data Elements (LDEs), wherein each of the LDEs is at least one of text data, linguistic data, image data, video data, sound data, control data, measurement data, olfactive data and tactile data, wherein identifying a first subset comprises;

retrieving, from the at least one storage device, the predetermined set of stored LDEs; and

comparing the unstructured data to the predetermined set of stored LDEs;

identifying a second subset of the plurality of data elements as Information Carrying Data Elements (ICDEs), wherein the second subset is comprised of every data element not identified as an LDE; and

representing the unstructured data in a structured format based on the identification of LDEs and ICDEs.

14. The computer controlled system of claim 13, wherein the method further comprises acts of:

identifying a plurality of LDE sequences within the first subset, each LDE sequence of the plurality of LDE sequences comprising at least one LDE; and

identifying a plurality of ICDE sequences within the second subset, each ICDE sequence of the plurality of ICDE sequences comprising at least one ICDE.

15. The computer controlled system of claim 14, wherein the act of representing the unstructured data comprises:

ranking the plurality of LDE sequences and/or the plurality of ICDE sequences based on a property of each of the sequences; and

representing the unstructured data in a structured format based on the act of ranking.

16. The computer controlled system of claim 15, wherein the property comprises at least one of:

a length associated with each of the sequences;

an order of occurrence of each of the sequences in the unstructured data;

a frequency of occurrence of each of the sequences in the unstructured data; and

one or more links to other sequences.

17. The computer controlled system of claim 14, wherein the method further comprises acts of:

identifying one or more subsequences within the plurality of LDE sequences and/or the plurality of ICDE sequences, wherein the one or more subsequences occur in at least two different sequences of the plurality of LDE sequences and/or the plurality of ICDE sequences.

18. The computer controlled system of claim 17, wherein the act of representing the unstructured data comprises:

ranking the plurality of LDE sequences and/or the plurality of ICDE sequences based on a property of the one or more subsequences; and

representing the unstructured data in a structured format based on the act of ranking.

20. The computer controlled system of claim 17, wherein the act of representing the unstructured data comprises:

ranking the plurality of LDE sequences and/or the plurality of ICDE sequences based on co-occurring subsequences; and

representing the unstructured data in a structured format based on the act of ranking.

21. The computer controlled system of claim 13, wherein the method further comprises acts of:

retrieving, from the repository, predefined information associated with at least one LDE of the predetermined set of stored LDEs and;

ranking the LDE sequences in the first subset based on the predefined information; and

wherein the act of representing comprises representing the unstructured data in a structured format based on the act of ranking.

22. The computer controlled system of claim 21, wherein the predefined information comprises a semantic denotation of the at least one LDE.

23. The computer controlled system of claim 21, wherein the predefined information comprises a qualification in a qualification system of the at least one LDE.

24. The computer controlled system of claim 23, wherein the qualification comprises at least one of:

an indication of mood, an indication of judgment on a positive/negative scale, and a number.