System, process and software arrangement for assisting with a knowledge discovery

• US 20040068476A1
• Filed: 10/20/2003
• Published: 04/08/2004
• Est. Priority Date: 01/04/2001
• Status: Active Grant

• Current Assignee: Abraham Bernstein, Foster Provost
• Original Assignee: Abraham Bernstein, Foster Provost
• Inventors: Bernstein, Abraham, Provost, Foster
• Granted Patent: US 7,167,846 B2
• Time in Patent Office: Days
• US Class Current: 706/45
• CPC Class Codes: G06N 5/01 Dynamic search techniques; Heuristics; Dynamic trees; Branch-and-bound
• CPC Class Codes: G06N 5/022 Knowledge engineering; Knowledge acquisition
• CPC Class Codes: G06N 5/025 Extracting rules from data
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks

Abstract

A process, system and computer software are provided to produce at least two solutions related to a knowledge discovery from data. In particular, the information regarding operators which are usable for the knowledge discovery of the data is received, and the solutions are generated. Each of the solutions includes at least one of the operators. An ability is provided to select at least one of the solutions so as to execute one or more procedures from the data. Each of the procedures is associated with the operator of the respective solution. In addition, it is possible to include a variable number of the operators in at least one (and possibly all) of the solutions. Also, it is possible to generate a code for at least one automatically-generated solution. In particular, access to the automatically-generated solution can be obtained with this solution that includes operators usable for the knowledge discovery of the data. Then, the code can be generated for associating one of the operators of the automatically-generated solution with another one of the operators of the automatically-generated solution.

Claims

1. A process for providing at least two solutions related to a knowledge discovery from data, comprising the steps of:

a) receiving information regarding operators which are usable for the knowledge discovery from the data;

b) generating the at least two solutions, each of the solutions including at least one of the operators; and

c) providing an ability to select at least one of the solutions so as to execute one or more procedures on the data, each of the one or more procedures being associated with the at least one operator of the respective solution.

2. The process according to claim 1, wherein the one or more procedures associated with the respective solution operator perform operations for mining the data.

3. The process according to claim 1, further comprising the steps of:

d) prior to step (a), identifying the data to be subjected to the knowledge discovery; and

e) extracting meta data from the identified data, the meta data providing characteristics of the identified data.

4. The process according to claim 3, wherein the characteristics of the identified data include types of attributes of the identified data.

5. The process according to claim 4, wherein the attributes indicate that the identified data are at least one of continuous, categorical and incomplete.

6. The process according to claim 3, further comprising the step of:

f) receiving particular information which identifies at least one of a type of the solutions to be generated, a desired result of the knowledge discovery and a type of the desired result to be provided.

7. The process according to claim 6, further comprising the step of:

g) prior to step (b), verifying a validity of each of the at least two solutions based on the characteristics of the meta data.

8. The process according to claim 7, wherein step (b) includes the substep of collecting the at least one operator for each of the solutions to produce a list of solutions.

9. The process according to claim 8, wherein the at least one operator is provided in an ontology structure of operators.

10. The process according to claim 9, further comprising the step of:

h) receiving specifications of the operators provided in the ontology structure, wherein the collecting substep is performed using the specifications.

11. The process according to claim 8, wherein the at least one operator of each of the solutions is a knowledge discovery operator.

12. The process according to claim 8, wherein the at least one operator of each of the solutions is a data mining operator.

13. The process according to claim 8, wherein at least one of the solutions includes at least two operators, and wherein step (c) includes the substep of executing the procedures associated with the operators of each selected one of the solutions.

14. The process according to claim 6, wherein step (b) includes the substep of producing a rank for each of the solutions based on at least one predetermined criterion.

15. The process according to claim 14, wherein the at least one predetermined criterion includes the particular information.

16. The process according to claim 14, further comprising the step of:

i) prior to step (b), receiving further information for the at least one predetermined criterion.

17. The process according to claim 16, wherein the further information includes weights, each of the weights being associated with a particular attribute of the selected solution.

18. The process according to claim 17, wherein the weights are provided by a user.

19. The process according to claim 17, further comprising the step of:

j) prior to step (b), generating a score for each of the solutions based on the weights.

20. The process according to claim 19, wherein the score is generated using at least one of a heuristic procedure and an automatic feedback procedure.

21. The process according to claim 20, wherein step (b) includes the substeps of:

assembling the solutions in a solution list, and

adding, to the solution list, each solution for which the score was generated.

22. The process according to claim 21, wherein the at least one valid operator is provided in an ontology structure of operators.

23. The process according to claim 22, further comprising the step of:

k) receiving specifications of the operators provided in the ontology structure, wherein the ranking substep is performed based on the specifications.

24. The process according to claim 14, wherein the ranking substep includes:

supplying example data to each of the solutions, and

obtaining the rank for each of the solutions based on at least one of output, accuracy and performance of the solutions.

25. The process according to claim 14, wherein step (b) further includes the substep of assembling the solutions based on the rank associated with each of the solutions.

26. The process according to claim 1, further comprising the step of:

l) generating at least one of an executable code and a source code for at least one of the solutions.

27. The process according to claim 26, wherein the at least one of the executable code and the source code are generated for a particular one of the solutions when it is selected.

28. The process according to claim 26, wherein the at least one of the executable code and the source code executes the respective procedure associated therewith on the data.

29. The process according to claim 1, further comprising the step of:

l) prior to step (b), obtaining a template which includes information regarding execution sequences of the operators, wherein the solutions are generated based on the information providing in the template.

30. The process according to claim 29, wherein the information of the template includes restrictions regarding an execution sequence order of each of the operators with respect to one another.

31. The process according to claim 1, wherein a number of the operators in the solutions is not fixed.

32. A process for providing at least one solution related to a knowledge discovery for data, comprising the steps of:

a) receiving information regarding operators which are usable for the knowledge discovery of the data; and

b) generating the at least one solution which includes a number of the operators, wherein the number of the operators in the solutions is not fixed.

33. The process according to claim 32, wherein the at least one solution includes a plurality of solutions, at least one of the solutions including at least two of the operators, and wherein the number of the operators of one of the solutions is capable of being different from the number of the operators of another one of the solutions.

34. The process according to claim 33, further comprising the step of:

c) providing an ability to select at least one of the solutions so as to execute one or more procedures on the data, each of the one or more procedures being associated with the at least one operator of the respective solution.

35. The process according to claim 34, wherein the operators are at least one of data mining operators and knowledge discovery operators.

36. A process for producing a code for at least one solution which is related to a knowledge discovery from data, comprising the steps of:

a) obtaining access to the at least one automatically-generated solution which includes operators that are usable for the knowledge discovery from the data; and

b) generating the code for associating one of the operators of the at least one automatically-generated solution with another one of the operators of the at least one automatically-generated solution.

37. A software system which, when executed on a processing device, provides at least two solutions related to a knowledge discovery from data, the software system comprising:

a processing subsystem which, when executed on the processing device, configures the processing device to perform the following;

a) receives information regarding operators which are usable for the knowledge discovery from the data,

b) generates the at least two solutions, each of the solutions including at least one of the operators, and

c) provides an ability to select at least one of the solutions so as to execute one or more procedures on the data, each of the one or more procedures being associated with the at least one operator of the respective solution.

38. The software system according to claim 37, wherein the one or more procedures associated with the respective solution operator perform operations for mining the data.

39. The software system according to claim 37, wherein, when executed on the processing device, the processing subsystem is further capable of:

prior to receiving the information, identifying the data to be subjected to the knowledge discovery, and

extracting meta data from the identified data, the meta data providing characteristics of the identified data.

40. The software system according to claim 39, wherein the characteristics of the identified data include types of attributes of the identified data.

41. The software system according to claim 40, wherein the attributes indicate that the identified data are at least one of continuous, categorical and incomplete.

42. The software system according to claim 38, wherein, when executed on the processing device, the processing subsystem is further capable of:

receiving particular information which identifies at least one of a type of the solutions to be generated, a desired result of the knowledge discovery and a type of the desired result to be provided.

43. The software system according to claim 42, wherein, when executed on the processing device, the processing subsystem is further capable of:

prior to the generation of the solutions, verifying a validity of each of the at least two solutions based on the characteristics of the meta data.

44. The software system according to claim 43, wherein the generation of the solutions is performed by collecting the at least one operator for each of the solutions to produce a list of solutions.

45. The software system according to claim 44, wherein the at least one operator is provided in an ontology structure of operators.

46. The software system according to claim 45, wherein, when executed on the processing device, the processing subsystem is further capable of:

receiving specifications of the operators provided in the ontology structure, wherein the collecting substep is performed using the specifications.

47. The software system according to claim 44, wherein the at least one operator of each of the solutions is a knowledge discovery operator.

48. The software system according to claim 44, wherein the at least one operator of each of the solutions is a data mining operator.

49. The software system according to claim 44, wherein at least one of the solutions includes at least two operators, and wherein the ability to select the at least one of the solutions includes a function to execute the procedures associated with the operators of each selected one of the solutions.

50. The software system according to claim 42, wherein the generation of the solutions include a procedure to produce a rank for each of the solutions based on at least one predetermined criterion.

51. The software system according to claim 50, wherein the at least one predetermined criterion includes the particular information.

52. The software system according to claim 50, wherein, when executed on the processing device, the processing subsystem is further capable of:

prior to the generation of the solutions, receiving further information for the at least one predetermined criterion.

53. The software system according to claim 52, wherein the further information includes weights, each of the weights being associated with a particular attribute of the selected solution.

54. The software system according to claim 53, wherein the weights are provided by a user.

55. The software system according to claim 53, wherein, when executed on the processing device, the processing subsystem is further capable of:

d) prior to the generation of the solutions, generating a score for each of the solutions based on the weights.

56. The software system according to claim 55, wherein the score is generated using at least one of a heuristic procedure and an automatic feedback procedure.

57. The software system according to claim 56, wherein the generation of the solutions includes:

assembling the solutions in a solution list, and

adding, to the solution list, each solution for which the score was generated.

58. The software system according to claim 57, wherein the at least one valid operator is provided in an ontology structure of operators.

59. The software system according to claim 58, wherein, when executed on the processing device, the processing subsystem is further capable of:

receiving specifications of the operators provided in the ontology structure, wherein the ranking substep is performed based on the specifications.

60. The software system according to claim 50, wherein the production of the ranks includes:

supplying example data to each of the solutions, and

obtaining the rank for each of the solutions based on at least one of output, accuracy and performance of the solutions.

61. The software system according to claim 50, wherein the generation of the solutions further includes assembling the solutions based on the rank associated with each of the solutions.

62. The software system according to claim 37, wherein, when executed on the processing device, the processing subsystem is further capable of:

generating at least one of an executable code and a source code for at least one of the solutions.

63. The software system according to claim 62, wherein the at least one of the executable code and the source code are generated for a particular one of the solutions when it is selected.

64. The software system according to claim 63, wherein the at least one of the executable code and the source code executes the respective procedure associated therewith on the data.

65. The software system according to claim 37, wherein, when executed on the processing device, the processing subsystem is further capable of:

prior to the generation of the solutions, obtaining a template which includes information regarding execution sequences of the operators, wherein the solutions are generated based on the information providing in the template.

66. The software system according to claim 65, wherein the information of the template includes restrictions regarding an execution sequence order of each of the operators with respect to one another.

67. The software system according to claim 37, wherein a number of the operators in the solutions is not fixed.

68. A software system which, when executed on a processing device, provides at least one solution related to a knowledge discovery for data, the software system comprising:

a processing subsystem which, when executed on the processing device, configures the processing device to perform the following;

a) receive information regarding operators which are usable for the knowledge discovery of the data, and

b) generate the at least one solution which includes a number of the operators, wherein the number of the operators in the at least one solution is not fixed.

69. The software system according to claim 68, wherein the at least one solution includes a plurality of solutions, at least one of the solutions including at least two of the operators, and wherein the number of the operators of one of the solutions is capable of being different from the number of the operators of another one of the solutions.

70. The software system according to claim 69, wherein, when executed on the processing device, the processing subsystem is further capable of:

providing an ability to select at least one of the solutions so as to execute one or more procedures on the data, each of the one or more procedures being associated with the at least one operator of the respective solution.

71. The software system according to claim 70, wherein the operators are at least one of data mining operators and knowledge discovery operators.

72. A software system which, when executed on the processing device, produces a code for at least one solution which is related to a knowledge discovery for data, comprising:

a processing subsystem which, when executed on the processing device, configures the processing device to perform the following;

a) obtain access to the at least one automatically-generated solution which includes operators that are usable for the knowledge discovery of the data, and

b) generate the code for associating one of the operators of the automatically-generated solution with another one of the operators of the at least one automatically-generated solution.

73. A computer software executable on a processing device to provide at least two solutions related to a knowledge discovery for data, the computer software comprising:

a) a first module which, when executed by the processing device, receives information regarding operators which are usable for the knowledge discovery of the data;

b) a second module which, when executed by the processing device, generates at least two solutions, each of the solutions including at least one of the operators; and

c) a third module which, when executed by the processing device, provides an ability to select at least one of the solutions so as to execute one or more procedures on the data, each of the one or more procedures being associated with the at least one operator of the respective solution.

74. A computer software executable on a processing device to provide at least one solution related to a knowledge discovery for data, the computer software comprising:

a) a first module which, when executed on the processing device, receives information regarding operators which are usable for the knowledge discovery of the data; and

b) a second module which, when executed on the processing device, generates the at least one solution which includes a variable number of the operators.

75. A computer software executable on a processing device to produce a code for at least one solution which is related to a knowledge discovery for data, the computer software comprises:

a) a first module which, when executed on the processing device, obtains access to the at least one automatically-generated solution which includes operators which are usable for the knowledge discovery of the data; and

b) a second module which, when executed on the processing device, generates the code for associating one of the operators of the at least one automatically-generated solution with another one of the operators of at least one automatically-generated solution.

76. A process for providing at least two solutions related to a knowledge discovery from data, comprising the steps of:

a) receiving information regarding operators which are usable for the knowledge discovery from the data;

b) generating the at least two solutions, each of the solutions including at least two of the operators; and

c) providing an ability to select at least one of the solutions so as to execute one or more procedures on the data, each of the procedures being associated with the at least two operators of the respective solution.

77. A software system which, when executed on a processing device, provides at least two solutions related to a knowledge discovery from data, the software system comprising:

a processing subsystem which, when executed on the processing device, configures the processing device to perform the following;

a) receives information regarding operators which are usable for the knowledge discovery from the data,

b) generates the at least two solutions, each of the solutions including at least two of the operators, and

c) provides an ability to select at least one of the solutions so as to execute one or more procedures on the data, each of the procedures being associated with the at least two operators of the respective solution.