Knowledge management using text classification

• US 20030172357A1
• Filed: 03/11/2002
• Published: 09/11/2003
• Est. Priority Date: 03/11/2002
• Status: Active Grant

• Current Assignee: The Boeing Co.
• Original Assignee: The Boeing Co.
• Inventors: Kao, Anne S.W., Woods, Steven D., Poteet, Stephen R., Quach, Lesley Hong-Minh
• Granted Patent: US 7,673,234 B2
• Time in Patent Office: Days
• US Class Current: 715/529
• CPC Class Codes: G06F 16/353 into predefined classes

Abstract

A method and system are provided for determining metadata to be assigned to a document. A document is identified to be published to a collection of documents. A category for publishing the document to the collection is automatically recommended using a text classifier. At least one category for the document may be selected, and the selected category is assigned as metadata for the document. A method and system are also provided for constructing a search for a document in a collection of documents. A search query is entered for a document in a collection of documents. A search category for searching for the document in the collection of documents is automatically recommended using a text classifier. At least one search category for searching for the document is selected.

Claims

1. A method for determining metadata to be assigned to a document, the method comprising:

identifying a document to be published to a collection of documents;

automatically recommending a knowledge category for publishing the document to the collection of documents using a text classifier;

selecting a knowledge category for the document; and

assigning the selected knowledge category as metadata for the document.

2. The method of claim 1, wherein automatically recommending a knowledge category further includes:

reading text of the document;

retrieving a stored set of knowledge categories;

associating the text of the document with the set of knowledge categories; and

recommending a knowledge category when association of the text of the document with the knowledge category exceeds a predetermined threshold.

3. The method of claim 2, wherein the text classifier is a Naï

ve Bayes text classifier.

4. The method of claim 2, wherein the text of the document is associated with the knowledge category when a numerical value for the association exceeds a predetermined probability that the knowledge category applies to the document.

5. The method of claim 2, wherein the text of the document is associated with the knowledge category when the text of the document includes the knowledge category.

6. The method of claim 1, wherein selecting a knowledge category is performed automatically.

7. The method of claim 1, wherein selecting a knowledge category is performed manually.

8. The method of claim 7, wherein the knowledge category selected for the document is recommended by the text classifier.

9. The method of claim 7, wherein the knowledge category selected for the document is not recommended by the text classifier.

10. The method of claim 1, further comprising expanding a knowledge category to at least one related knowledge category.

11. A method for constructing a query for at least one document in a collection of documents, the method comprising:

entering a first query for at least one document in a collection of documents;

automatically recommending a knowledge category for constructing a knowledge category query; and

selecting a knowledge category for the knowledge category query.

12. The method of claim 11, further comprising submitting the knowledge category query with the selected knowledge category.

13. The method of claim 12, further comprising automatically returning at least one document that includes text that is associated with the knowledge category query submitted.

14. The method of claim 11, wherein the first query is a natural language query.

15. The method of claim 11, wherein automatically recommending a knowledge category further includes:

reading the first query;

retrieving a stored set of knowledge categories;

associating the first query with the set of knowledge categories; and

recommending a knowledge category when association of the first query with the knowledge category exceeds a predetermined threshold.

16. The method of claim 15, further comprising submitting the knowledge category query with the selected knowledge category.

17. The method of claim 15, wherein the first query is associated with the knowledge category when a numerical value for the association exceeds a predetermined probability that the knowledge category applies to the first query.

18. The method of claim 15, wherein the first query is associated with the knowledge category when the first query includes the knowledge category.

19. The method of claim 11, wherein the knowledge category selected for the knowledge category query is recommended by the text classifier.

20. The method of claim 11, wherein the knowledge category selected for the knowledge category query is not recommended by the text classifier.

21. The method of claim 11, further comprising expanding a knowledge category to at least one related knowledge category.

22. A system for determining metadata to be assigned to a document, the system comprising:

means for identifying a document to be published to a collection of documents;

means for automatically recommending a knowledge category for publishing the document to the collection of documents using a text classifier;

means for selecting a knowledge category for the document; and

means for assigning the selected knowledge category as metadata for the document.

23. The system of claim 22, wherein the automatic knowledge category recommending means further includes:

means for reading text of the document;

means for retrieving a stored set of knowledge categories;

means for associating the text of the document with the set of knowledge categories; and

means for recommending a knowledge category when association of the text of the document with the knowledge category exceeds a predetermined threshold.

24. The system of claim 23, wherein the text classifier is a Naï

ve Bayes text classifier.

25. The system of claim 23, wherein the text of the document is associated with the knowledge category when a numerical value for the association exceeds a predetermined probability that the knowledge category applies to the document.

26. The system of claim 23, wherein the text of the document is associated with the knowledge category when the text of the document includes the knowledge category.

27. The system of claim 22, wherein the knowledge category is automatically selected.

28. The system of claim 22, wherein the knowledge category is manually selected.

29. The system of claim 28, wherein the knowledge category selected for the document is recommended by the text classifier.

30. The system of claim 28, wherein the knowledge category selected for the document is not recommended by the text classifier.

31. The system of claim 22, further comprising means for expanding a knowledge category to at least one related knowledge category.

32. A system for constructing a query for at least one document in a collection of documents, the system comprising:

means for entering a first query for at least one document in a collection of documents;

means for automatically recommending a knowledge category for constructing a knowledge category query; and

means for selecting a knowledge category for the knowledge category query.

33. The system of claim 32, further comprising means for submitting the knowledge category query with the selected knowledge category.

34. The system of claim 33, further comprising means for automatically returning at least one document that includes text that is associated with the knowledge category query submitted.

35. The system of claim 32, wherein the first query is a natural language query.

36. The system of claim 32, wherein the automatic knowledge category recommending means further includes:

means for reading the first query;

means for retrieving a stored set of knowledge categories;

means for associating the first query with the set of knowledge categories; and

means for recommending a knowledge category when association of the first query with the knowledge category exceeds a predetermined threshold.

37. The system of claim 36, further comprising means for submitting the knowledge category query with the selected knowledge category.

38. The system of claim 36, wherein the first query is associated with the knowledge category when a numerical value for the association exceeds a predetermined probability that the knowledge category applies to the first query.

39. The system of claim 36, wherein the first query is associated with the knowledge category when the first query includes the knowledge category.

40. The system of claim 32, wherein the knowledge category selected for the knowledge category query is recommended by the text classifier.

41. The system of claim 32, wherein the knowledge category selected for the knowledge category query is not recommended by the text classifier.

42. The system of claim 32, further comprising means for expanding a knowledge category to at least one related knowledge category.

43. A computer program product for determining metadata to be assigned to a document, the computer program product comprising:

first computer-readable program code means for identifying a document to be published to a collection of documents;

second computer-readable program code means for automatically recommending a knowledge category for publishing the document to the collection of documents using a text classifier;

third computer-readable program code means for selecting a knowledge category for the document; and

fourth computer-readable program code means for assigning the selected knowledge category as metadata for the document.

44. The computer program product of claim 43, wherein the second computer-readable program code means for automatically recommending a knowledge category further includes:

fifth computer-readable program code means for reading text of the document;

sixth computer-readable program code means for retrieving a stored set of knowledge categories;

seventh computer-readable program code means for associating the text of the document with the set of knowledge categories; and

eighth computer-readable program code means for recommending a knowledge category when association of the text of the document to the knowledge category exceeds a predetermined threshold.

45. The computer program product of claim 44, wherein the text classifier is a Naï

ve Bayes text classifier.

46. The computer program product of claim 44, wherein the text of the document is associated with the knowledge category when a numerical value for the association exceeds a predetermined probability that the knowledge category applies to the document.

47. The computer program product of claim 44, wherein the text of the document is associated with the knowledge category when the text of the document includes the knowledge category.

48. The computer program product of claim 43, wherein the knowledge category is automatically selected.

49. The computer program product of claim 43, wherein the knowledge category is manually selected.

50. The computer program product of claim 49, wherein the knowledge category selected for the document is recommended by the text classifier.

51. The computer program product of claim 49, wherein the knowledge category selected for the document is not recommended by the text classifier.

52. The computer program product of claim 43, further comprising ninth computer-readable program code means for expanding a knowledge category to at least one related knowledge category.

53. A computer program product for constructing a query for at least one document in a collection of documents, the computer program product comprising:

first computer-readable program code means for entering a first query for at least one document in a collection of documents;

second computer-readable program code means for automatically recommending a knowledge category for constructing a knowledge category query; and

third computer-readable program code means for selecting a knowledge category for the knowledge category query.

54. The computer program product of claim 53, further comprising fourth computer-readable program code means for submitting the knowledge category query with the selected knowledge category.

55. The computer program product of claim 54, further comprising fifth computer-readable program code means for automatically returning at least one document that includes text that is associated with the knowledge category query submitted.

56. The computer program product of claim 53, wherein the first query is a natural language query.

57. The computer program product of claim 53, wherein the second computer-readable program code means for automatically recommending a knowledge category further includes:

sixth computer-readable program code means for reading the first query;

seventh computer-readable program code means for retrieving a stored set of knowledge categories;

eighth computer-readable program code means for associating the first query with the set of knowledge categories; and

ninth computer-readable program code means for recommending a knowledge category when association of the first query with the knowledge category exceeds a predetermined threshold.

58. The computer program product of claim 57, further comprising tenth computer-readable program code means for submitting the knowledge category query with the selected knowledge category.

59. The computer program product of claim 57, wherein the first query is associated with the knowledge category when a numerical value for the association exceeds a predetermined probability that the knowledge category applies to the first query.

60. The computer program product of claim 57, wherein the first query is associated with the knowledge category when the first query includes the knowledge category.

61. The computer program product of claim 53, wherein the knowledge category selected for the knowledge category query is recommended by the text classifier.

62. The computer program product of claim 53, wherein the knowledge category selected for the knowledge category query is not recommended by the text classifier.

63. The computer program product of claim 53, further comprising eleventh computer-readable program code means for expanding a knowledge category to at least one related knowledge category.