Knowledge Discovery Appartus and Method

• US 20050038805A1
• Filed: 08/12/2003
• Published: 02/17/2005
• Est. Priority Date: 08/12/2003
• Status: Active Grant

• Current Assignee: Viziant Corp.
• Original Assignee: Eagleforce Associates Incorporated
• Inventors: Campbell, Stanley V., Maren, Alianna J.
• Granted Patent: US 7,333,997 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/435 Filtering based on additional data, e.g. user or group profiles
• CPC Class Codes: G06F 2216/03 Data mining
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation

Abstract

A knowledge discovery apparatus and method that extracts both specifically desired as well as pertinent and relevant information to query from a corpus of multiple elements that can be structured, unstructured, and/or semi-structured, along with imagery, video, speech, and other forms of data representation, to generate a set of outputs with a confidence metric-applied to the match of the output against the query. The invented apparatus includes a multi-level architecture, along with one or more feedback loop(s) from any Level N to any lower Level so that a user can control the output of this knowledge discovery method via providing inputs to the utility function.

Claims

1-16. -16:

CANCELLED

17. A system for identifying elements in a dataset including structured, semi-structured or unstructured data comprising a plurality of elements from a plurality of data sources, wherein each of the plurality of elements includes at least one corresponding metatag indicative of content of the corresponding element, comprising:

a plurality of processing levels, wherein each processing level includes

a processor configured to identify a group of elements that satisfy a set of criteria; and

a filter configured to extract a subset of the group of elements identified by the processor that satisfy a selection criteria and provide the subset of elements to a next processing level;

a feedback component configured to provide a feedback loop between two of the processing levels, wherein the feedback component is configured to provide a portion of the group of elements identified by one of the processing levels to another one of the processing levels so that the processing level receiving the portion of the group of elements identifies a new group of elements contained in the portion of the group of elements provided by the feedback component.

18. The system of claim 17, further comprising:

a utility component configured to control the feedback component, wherein the utility component is configured to determine which portion of the elements are provided to a particular one of the processing levels.

19. The system of claim 18, wherein the utility component is configured to determine the number of times the feedback component provides the portion of the group of elements to the particular processing level.

20. A system for identifying elements in a dataset including structured, semi-structured unstructured data comprising a plurality of elements from a plurality of data sources, wherein each of the plurality of elements includes at least one corresponding metatag indicative of content of the corresponding element, comprising:

a plurality of processors, wherein each processor is configured to identify a group of elements that satisfy a set of criteria;

a plurality of filters, wherein each filter is configured to extract a subset of the group of elements identified by one of the processors that satisfy a selection criteria and provide the subset of elements to a next one of the plurality of processors;

a feedback component configured to provide a feedback loop between two of the plurality of processors, wherein the feedback component is configured to provide a portion of the group of elements identified by one of the plurality of processors to another one of the plurality of processors such that the processor receiving the portion of the group of elements identifies a new group of elements contained in the portion of the group of elements provided by the feedback component.

21. The system of claim 20, wherein the feedback component is configured to provide feedback loops between more than two of the plurality of processors.

22. The system of claim 20, wherein a different set of criteria is employed by each of the plurality of processors.

23. The system of claim 20, wherein the selection criteria employed by each of the plurality of filters is different.

24. The system of claim 20, wherein the selection criteria is different than the set of criteria.

25. The system of claim 20, further comprising:

a utility component configured to control the feedback component, wherein the utility component is configured to determine which portion of the elements are provided to a particular one of the plurality of processors.

26. The system of claim 25, wherein the utility component is configured to determine the number of times the feedback component provides the portion of the group of elements to the particular processor.

27. The system of claim 23, further comprising:

a tagging processor configured to assign one or more metatags to each of the elements in the database.

28. The system of claim 20, wherein the set of criteria employed by each of the processors either set by a user or set automatically.

29. The system of claim 20, wherein the set of criteria employed by at least one of the processors is based on the pairwise association of the data elements.

30. The system of claim 29, wherein the pairwise associations are based on nouns or noun phrases among the elements of the group.

31. The system of claim 30, wherein at least one of the processors generates a pairwise occurrence matrix, wherein each element of the matrix being incremented when a pair of nouns or noun phrases occur within a set distance of each other.

32. The system of claim 20, wherein the selection criteria includes establishing a threshold that is used to selects elements of the matrix that have been incremented to at least equal the value of the threshold.

33. The system of claim 20, wherein the set of criteria employed by at least one of the processors is based on syntactic associations of elements in the dataset.

34. The system of claim 33, wherein the syntactic associations are based on noun-verb associations and verb-object noun associations among the elements of one of the subsets extracted by one of the plurality of filters.