Database system with query relaxation using type abstraction hierarchy (TAH) as query condition relaxation structure

• US 5,956,707 A
• Filed: 02/13/1997
• Issued: 09/21/1999
• Est. Priority Date: 02/13/1997
• Status: Expired due to Fees

• Current Assignee: Wesley W. Chu
• Original Assignee: Wesley W. Chu
• Inventors: Chu, Wesley W.
• Primary Examiner(s): Kulik, Paul V.
• Assistant Examiner(s): Corrielus, Jean M.
• Application Number:
  US08/800,035
• Time in Patent Office: 950 Days
• Field of Search: 364/283.15, 364/974, 364/274.14, 395/603, 395/12, 395/10, 395/1, 707/1, 707/3
• US Class Current: 707/759
• CPC Class Codes: G06F 16/2452 Query translation
• CPC Class Codes: G06F 16/24573 using data annotations, e.g. user-defined metadata
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/2468 Fuzzy queries
• CPC Class Codes: Y10S 707/99931 Database or file accessing
• CPC Class Codes: Y10S 707/99933 Query processing, i.e. searching

Abstract

A system for obtaining target data from a conventional computer database in response to an input query which has at least one attribute includes a Type Abstraction Hierarchy (TAH) manager for providing a TAH structure including relaxation conditions for said at least one attribute. A control unit successively applies a database query which corresponds to the input query to and receives data from the database, with progressively relaxed conditions of the attribute(s) being provided by the TAH manager, until the target data in the form of a specified number of ranked answers which satisfy the attribute conditions is obtained. The attribute relaxation process can be controlled such as relaxation order, preference list, reject, etc. to obtain user and context specific answers. The system can produce approximate answers or answers to query with conceptual terms. The input query can including cooperative operators such as "APPROXIMATELY", "NEAR TO" or "SIMILAR TO" or conceptual terms (e.g. "long"), which the conventional database is not capable of directly processing. The control unit converts the cooperative operators and conceptual terms into database operator which the database is capable of processing. The TAH structure includes TAHs having leaves corresponding to instances of the attributes respectively, and a hierarchical arrangement of nodes that specify ranges which include at least one of the instances respectively. The instances can be numerical and/or non-numerical. The nodes can be defined by conceptual names (terms) that describe the instances under that node (e.g. dark color, long runway).

Claims

1. A system for obtaining target data from a computer database in response to an input query which includes at least one attribute, comprising:

a Type Abstraction Hierarchy (TAH) manager for providing a TAH structure, including relaxation conditions for said at least one attribute; and

a control unit for successively applying a database query which corresponds to the input query to and receiving data from the database, with progressively relaxed conditions of said at least one attribute provided by the TAH manager, until the targer data in the form of at least one answer which satisfies said at least one attribute is obtained.

2. A system as in claim 1, in which:

the input query can include a cooperative operator;

the database is not capable of directly processing the cooperative operator; and

the control unit converts the cooperative operator into a database operator which the database is capable of processing.

3. A system as in claim 2, in which the cooperative operator is one of a value-based operator, a condition-based operator, and a relaxation control operator.

4. A system as in claim 3, in which the condition-based operator is SIMILAR-TO with a clause that BASED-ON at least one attribute.

5. A system as in claim 3, in which the relaxation control operator comprises at least one of relaxation order, preference list, relaxation level, reject, unrelaxable, at least, and use TAHx, where x is the name of a TAH to be used for relaxation.

6. A system as in claim 2, in which the control unit comprises:

a query parser for converting the input query into a query representation which the TAH manager is capable of processing;

a preprocessor for converting the cooperative operator in the query representation into the database operator; and

a data source manager for converting the query representation into the database query, and applying the database query to the database.

7. A system as in claim 1, in which the control unit comprises a relaxation unit for accessing the TAH manager in accordance with a specified procedure.

8. A system as in claim 7, in which the relaxation unit comprises:

a relaxation controller for accessing the TAH manager; and

a condition modifier for modifying said at least one attribute in accordance with progressively relaxed conditions thereof returned by the TAH manager.

9. A system as in claim 8, in which the control unit comprises:

a data source manager for applying the database query to and obtaining data from the database; and

a controller for controlling operation of the data source manager, relaxation unit, and TAH manager.

10. A system as in claim 7, in which:

said at least one attribute comprises a plurality of attributes; and

said specified procedure comprises relaxing said plurality of attributes in a specified order.

11. A system as in claim 10, in which the input query comprises specification of said order.

12. A system as in claim 1, in which:

the input query is specific; and

the target data comprises an approximate answer to the input query.

13. A system as in claim 1, in which the input query contains conceptual terms.

14. A system as in claim 1, in which the TAH structure comprises at least one TAH for each attribute respectively.

15. A system as in claim 1, in which the TAH structure comprises a TAH having leaves corresponding to instances of one of said at least one attribute respectively, and a hierarchical arrangement of nodes that specify ranges which include at least one of said instances respectively.

16. A system as in claim 15, in which the instances are numerical.

17. A system as in claim 15, in which the instances are non-numerical.

18. A system as in claim 15, in which the nodes have conceptual names.

19. A system as in claim 1, in which:

said at least one attribute comprises a plurality of attributes; and

the TAH structure comprises a multiple attribute TAH which includes relaxation conditions for at least two of said plurality of attributes.

20. A system as in claim 19, in which said relaxation conditions are such that said at least two of said plurality of attributes in the multiple attribute TAH are relaxed together.

21. A system as in claim 1, in which:

said at least one attribute comprises a plurality of attributes; and

the TAH structure comprises a first multiple attribute TAH which includes relaxation conditions for at least two of said plurality of attributes; and

a second multiple attribute TAH which includes relaxation conditions for at least another two of said plurality of attributes.

22. A system as in claim 21, in which said relaxation conditions are such that said at least two of said plurality of attributes in the first multiple attribute TAH are relaxed together;

and said at least another two of said plurality of attributes in the second multiple attribute TAH are relaxed together.

23. A system as in claim 22, in which said relaxation conditions are such that the first multiple attribute TAH and the second multiple attribute TAH are relaxed in a specified order.

24. A system as in claim 1, in which:

the target data comprises a plurality of answers; and

the control unit further comprises a postprocessor for ranking the answers in accordance with a specified criterion.

25. A system as in claim 24, in which the input query specifies said criterion.

26. A system as in claim 1, in which:

the database comprises a schema including said at least one attribute;

the control unit comprises a schema manager which accesses the database to obtain a specification of the schema; and

the control unit further comprises a query parser which checks syntax of the input query by obtaining said specification of the schema from the schema manager.

27. A system as in claim 1, in which the TAH manager comprises means for editing the TAH structure.

28. A system as in claim 1, in which the TAH manager comprises means for automatically obtaining information from the database and creating the TAH structure in accordance therewith.

29. A computer implemented method for obtaining target data from a computer database in response to an input query which includes at least one attribute, comprising the steps of:

(a) providing a Type Abstraction Hierarchy (TAH) structure including relaxation conditions for said at least one attribute;

(b) applying a database query which corresponds to the input query to the database;

(c) receiving data from the database;

(d) if the data received in step (c) comprises the target data, outputting the data and ending the method;

(e) if the data received in step (c) does not comprise the target data, accessing the TAH structure to obtain a relaxed condition of said at least one attribute;

(f) modifying the database query to include said relaxed condition; and

(g) jumping to step (c).

30. A method as in claim 29, in which:

the input query can include a cooperative operator;

the database is not capable of directly processing the cooperative operator; and

the method further comprises the step, performed prior to step (b), of;

(i) converting the cooperative operator into a database operator which the database is capable of processing.

31. A method as in claim 30, in which the cooperative operator is one of a value-based operator, a condition-based operator, and a relaxation control operator.

32. A method as in claim 31, in which the condition-based operator is SIMILAR-TO, BASED-ON.

33. A system as in claim 31, in which the relaxation control operator comprises at least one of relaxation order, preference list, relaxation level, reject, unrelaxable, at least, and use TAHx, where x is a number of a TAH.

34. A method as in claim 29, in which step (e) comprises the substeps of:

(e1) accessing the TAH structure in accordance with a specified procedure to obtain said relaxed condition; and

(e2) modifying said at least one attribute to include said relaxed condition.

35. A method as in claim 34, in which:

said at least one attribute comprises a plurality of attributes; and

said specified procedure in step (e1) comprises relaxing said plurality of attributes in a specified order.

36. A method as in claim 35, in which the input query comprises specification of said order.

37. A method as in claim 29, in which:

the input query is specific; and

the target data comprises an approximate answer to the input query.

38. A method as in claim 29, in which the input query contains conceptual terms.

39. A method as in claim 29, in which the TAH structure comprises at least one TAH for each attribute respectively.

40. A method as in claim 29, in which the TAH structure comprises a TAH having leaves corresponding to instances of one of said at least one attribute respectively, and a hierarchical arrangement of nodes that specify ranges which include at least one of said instances respectively.

41. A method as in claim 40, in which the instances are numerical.

42. A method as in claim 29, in which the instances are non-numerical.

43. A method as in claim 29, in which at least some of the nodes have conceptual names.

44. A method as in claim 29, in which:

said at least one attribute comprises a plurality of attributes; and

the TAH structure comprises a multiple attribute TAH which includes relaxation conditions for at least two of said plurality of attributes.

45. A method as in claim 44, in which said relaxation conditions are such that said at least two of said plurality of attributes in the multiple attribute TAH are relaxed together.

46. A method as in claim 29, in which:

said at least one attribute comprises a plurality of attributes; and

the TAH structure comprises a first multiple attribute TAH which includes relaxation conditions for at least two of said plurality of attributes; and

a second multiple attribute TAH which includes relaxation conditions for at least another two of said plurality of attributes.

47. A method as in claim 46, in which said relaxation conditions are such that said at least two of said plurality of attributes in the first multiple attribute TAH are relaxed together;

and said at least another two of said plurality of attributes in the second multiple attribute TAH are relaxed together.

48. A method as in claim 47, in which said relaxation conditions are such that the first multiple attribute TAH and the second multiple attribute TAH are relaxed in a specified order.

49. A method as in claim 29, in which:

the target data comprises a plurality of answers; and

the method further comprises the step of;

(h) ranking the answers in accordance with a specified criterion.

50. A method as in claim 29, in which the input query specifies said criterion.

51. A storage medium for storing a computer program for obtaining target data from a database in response to an input query which includes at least one attribute, the program controlling a computer to perform the steps of:

(a) providing a Type Abstraction Hierarchy (TAH) structure including relaxation conditions for said at least one attribute;

(b) applying a database query which corresponds to the input query to the database;

(c) receiving data from the database;

(d) if the data received in step (c) comprises the target data, outputting the data and ending the method;

(e) if the data received in step (c) does not comprise the target data, accessing the TAH structure to obtain a relaxed condition of said at least one attribute;

(f) modifying the database query to include said relaxed condition; and

(g) jumping to step (c).

52. A medium as in claim 51, in which:

the input query can include a cooperative operator;

the database is not capable of directly processing the cooperative operator; and

the program controls the computer to perform the further step, prior to step (b), of;

(i) converting the cooperative operator into a database operator which the database is capable of processing.

53. A medium as in claim 52, in which the cooperative operator is one of a value-based operator, a condition-based operator, and a relaxation control operator.

54. A medium as in claim 53, in which the condition-based operator is SIMILAR-TO, BASED-ON.

55. A medium as in claim 53, in which the relaxation control operator comprises at least one of relaxation order, preference list, relaxation level, reject, unrelaxable, at least, and use TAHx, where x is a number of a TAH.

56. A medium as in claim 51, in which step (e) comprises the substeps of:

(e1) accessing the TAH structure in accordance with a specified procedure to obtain said relaxed condition; and

(e2) modifying said at least one attribute to include said relaxed condition.

57. A medium as in claim 56, in which:

said at least one attribute comprises a plurality of attributes; and

said specified procedure in step (e1) comprises relaxing said plurality of attributes in a specified order.

58. A medium as in claim 51, in which the input query comprises specification of said order.

59. A medium as in claim 51, in which:

the input query is specific; and

the target data comprises an approximate answer to the input query.

60. A medium as in claim 51, in which the input query is conceptual.

61. A medium as in claim 51, in which the TAH structure comprises at least one TAH for each attribute respectively.

62. A medium as in claim 51, in which the TAH structure comprises a TAH having leaves corresponding to instances of one of said at least one attribute respectively, and a hierarchical arrangement of nodes that specify ranges which include at least one of said instances respectively.

63. A medium as in claim 62, in which the instances are numerical.

64. A medium as in claim 62, in which the instances are non-numerical.

65. A medium as in claim 62, in which at least some of the nodes have conceptual names.

66. A medium as in claim 51, in which:

said at least one attribute comprises a plurality of attributes; and

the TAH structure comprises a multiple attribute TAH which includes relaxation conditions for at least two of said plurality of attributes.

67. A medium as in claim 66, in which said relaxation conditions are such that said at least two of said plurality of attributes in the multiple attribute TAH are relaxed together.

68. A medium as in claim 51, in which:

said at least one attribute comprises a plurality of attributes; and

the TAH structure comprises a first multiple attribute TAH which includes relaxation conditions for at least two of said plurality of attributes; and

a second multiple attribute TAH which includes relaxation conditions for at least another two of said plurality of attributes.

69. A medium as in claim 68, in which said relaxation conditions are such that said at least two of said plurality of attributes in the first multiple attribute TAH are relaxed together;

and said at least another two of said plurality of attributes in the second multiple attribute TAH are relaxed together.

70. A medium as in claim 69, in which said relaxation conditions are such that the first multiple attribute TAH and the second multiple attribute TAH are relaxed in a specified order.

71. A medium as in claim 51, in which:

the target data comprises a plurality of answers; and

the program controls the computer to perform the further step of;

(h) ranking the answers in accordance with a specified criterion.

72. A medium as in claim 71, in which the input query specifies said criterion.

73. A medium as in claim 51, comprising a magnetic storage disk.

74. A medium as in claim 51, comprising an optical storage disk.

75. A medium as in claim 51, comprising a computer memory.