Computer systems and methods for automatic generation of models for a dataset

• US 20070250523A1
• Filed: 04/17/2007
• Published: 10/25/2007
• Est. Priority Date: 04/19/2006
• Status: Active Grant

• Current Assignee: Tableau Software Incorporated (Salesforce.com, Inc.)
• Original Assignee: Andrew Beers, Jonathan Taylor, Matthew Eldridge, Patrick Hanrahan
• Inventors: Hanrahan, Patrick, Beers, Andrew, Taylor, Jonathan, Eldridge, Matthew
• Granted Patent: US 7,999,809 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/248 Presentation of query results
• CPC Class Codes: G06F 16/283 Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP
• CPC Class Codes: G06F 3/0481 based on specific properties of the displayed interaction object or a metaphor-based environment, e.g. interaction with desktop elements like windows or icons, or assisted by a cursor's changing behaviour or appearance
• CPC Class Codes: G06F 30/00 Computer-aided design [CAD]
• CPC Class Codes: G16B 40/00 ICT specially adapted for biostatistics; ICT specially adapted for bioinformatics-related machine learning or data mining, e.g. knowledge discovery or pattern finding

Abstract

A method of automatically generating models from a dataset includes multiple steps. First, a description of a view of a dataset is provided. The description includes multiple fields associated with the dataset. Next, a set of properties is determined for each of the multiple fields. Finally, the description is automatically translated into one or more models based on the respective properties of the multiple fields and a set of predefined heuristics.

Claims

1. A method of automatically generating models for a dataset, comprising:

generating a description to visualize a dataset, the description including multiple fields associated with a view of the dataset;

determining a set of properties for each of the multiple fields; and

automatically translating the description into one or more models based on the respective properties of the multiple fields and a set of predefined heuristics.

2. The method of claim 1, wherein the multiple fields are partitioned into a plurality of sets, each set associated with a respective region of the view including row, column, page, and an encoding area.

3. The method of claim 2, wherein each field includes a field type selected from the group consisting of categorical and quantitative and a data type selected from the group consisting of integer, real, Boolean, and time.

4. The method of claim 3, wherein the set of predefined heuristics includes selecting two fields from the description, which may be potentially related, as an independent variable and a response variable of the models, and converting categorical fields in the description into factors of the models.

5. The method of claim 4, wherein the set of predefined heuristics further comprises:

defining a first quantitative field on the row (or the column) as a response variable and a second field on the column (or the row) as an independent variable if the first quantitative field is the only quantitative field on the row or the column of the view and the second field is either a quantitative field or a categorical field having a quantitative interpretation; and

generating the one or more models using data in the dataset associated with the response variable and the independent variable.

6. The method of claim 4, wherein the set of predefined heuristics further comprises:

defining a first quantitative field on the row (or the column) as a response variable and a second quantitative field on the column (or the row) as an independent variable if the second quantitative field is a quantitative field from a dimension class; and

generating the one or more models using data in the dataset associated with the response variable and the independent variable.

7. The method of claim 4, wherein the set of predefined heuristics further comprises:

defining a first quantitative field on the row as a response variable and a second quantitative field on the column as an independent variable; and

generating the one or more models using data in the dataset associated with the response variable and the independent variable.

8. The method of claim 7, wherein the set of predefined heuristics further comprises:

repeating said defining and generating steps for each unique pair of first and second quantitative fields on the column and row.

9. The method of claim 4, wherein the set of predefined heuristics further comprises:

defining each categorical field on the row, the column and the page of the view as a categorical factor of the one or more models.

10. The method of claim 4, wherein the set of predefined heuristics further comprises:

defining each categorical field in the encoding area as a categorical factor of the one or more models if there are more than one data points corresponding to each value in the categorical field.

11. The method of claim 4, wherein the set of predefined heuristics further comprises:

ignoring each quantitative field in the encoding area of the view when translating the description into the one or more models.

12. The method of claim 1, wherein each of the one or more models has a model type including linear, polynomial, exponential, logarithmic, and transcendental.

13. The method of claim 1, further comprising:

displaying the view of the dataset in accordance with the description; and

displaying graphs corresponding to the one or more models.

14. The method of claim 1, further comprising:

determining if it is possible to translate the description into any model by examining the properties of the fields associated with the view.

15. The method of claim 1, wherein the one or more models are statistical models.

16. The method of claim 1, further comprising:

generating a plurality of models using the description, each model having a specific model type and a set of parameters;

automatically selecting from the plurality of models one having a simpler model type and fewer parameters according to predefined criteria.

17. The method of claim 1, wherein the description is generated by a user dragging and dropping a subset of the fields into respective field containers of a graphical user interface.

18. The method of claim 1, wherein the description is generated by a user selecting a subset of the fields as fields of interest through a graphical user interface.

19. The method of claim 1, wherein the description is generated by a user entering a text string including a subset of the fields as fields of interest through a user interface.

20. A method of automatically generating models for a dataset, comprising:

allowing a user to provide a description by dragging and dropping fields into field containers of a graphical user interface, the description including at least two sets of fields associated with a view of the dataset;

determining a set of properties for each of the two sets of fields;

generating one or more models based on the properties of the two sets of fields and a set of predefined heuristics; and

automatically plotting the one or more models in a specific region of the graphical user interface.

21. The method of claim 20, wherein the multiple fields are partitioned into a plurality of sets, each set associated with a respective region of the view including row, column, page, and an encoding area.

22. The method of claim 21, wherein each field includes a field type selected from the group consisting of categorical and quantitative and a data type selected from the group consisting of integer, real, Boolean, and time.

23. The method of claim 22, wherein the set of predefined heuristics includes selecting two fields from the description, which may be potentially related, as an independent variable and a response variable of the models, and converting categorical fields in the description into factors of the models.

24. The method of claim 23, wherein the set of predefined heuristics further comprises:

defining a first quantitative field on the row (or the column) as a response variable and a second field on the column (or the row) as an independent variable if the first quantitative field is the only quantitative field on the row or the column of the view and the second field is either a quantitative field or a categorical field having a quantitative interpretation; and

generating the one or more models using data in the dataset associated with the response variable and the independent variable.

25. The method of claim 23, wherein the set of predefined heuristics further comprises:

defining a first quantitative field on the row (or the column) as a response variable and a second quantitative field on the column (or the row) as an independent variable if the second quantitative field is a quantitative field from a dimension class; and

generating the one or more models using data in the dataset associated with the response variable and the independent variable.

26. The method of claim 23, wherein the set of predefined heuristics further comprises:

defining a first quantitative field on the row as a response variable and a second quantitative field on the column as an independent variable; and

generating the one or more models using data in the dataset associated with the response variable and the independent variable.

27. The method of claim 26, wherein the set of predefined heuristics further comprises:

repeating said defining and generating steps for each unique pair of first and second quantitative fields on the column and row.

28. The method of claim 23, wherein the set of predefined heuristics further comprises:

defining each categorical field on the row, the column and the page of the view as a categorical factor of the one or more models.

29. The method of claim 23, wherein the set of predefined heuristics further comprises:

defining each categorical field in the encoding area as a categorical factor of the one or more models if there are more than one data points corresponding to each value in the categorical field.

30. The method of claim 23, wherein the set of predefined heuristics further comprises:

ignoring each quantitative field in the encoding area of the view when translating the description into the one or more models.

31. The method of claim 20, wherein each of the one or more models has a model type including linear, polynomial, exponential, logarithmic, and transcendental.

32. The method of claim 20, further comprising:

displaying the view of the dataset in accordance with the description; and

displaying graphs corresponding to the one or more models.

33. The method of claim 20, further comprising:

determining if it is possible to translate the description into any model by examining the properties of the fields associated with the view.

34. The method of claim 20, wherein the one or more models are statistical models.

35. The method of claim 20, further comprising:

generating a plurality of models using the description, each model having a specific model type and a set of parameters;

automatically selecting from the plurality of models one having a simpler model type and fewer parameters according to predefined criteria.

36. The method of claim 20, wherein the description is generated by a user dragging and dropping a subset of the fields into respective field containers of a graphical user interface.

37. The method of claim 20, wherein the description is generated by a user selecting a subset of the fields as fields of interest through a graphical user interface.

38. The method of claim 20, wherein the description is generated by a user entering a text string including a subset of the fields as fields of interest through a user interface.

39. A computer system, comprising:

a main memory;

a processor; and

at least one program, stored in the main memory and executed by the processor, the at least one program further including;

instructions for generating a description to visualize a dataset, the description including multiple fields associated with a view of the dataset;

instructions for determining a set of properties for each of the multiple fields; and

instructions for automatically translating the description into one or more models based on the respective properties of the multiple fields and a set of predefined heuristics.

40. The computer system of claim 39, wherein the multiple fields are partitioned into a plurality of sets, each set associated with a respective region of the view including row, column, page, and an encoding area.

41. The computer system of claim 40, wherein each field includes a field type selected from the group consisting of categorical and quantitative and a data type selected from the group consisting of integer, real, Boolean, and time.

42. The computer system of claim 41, wherein the set of predefined heuristics includes instructions for selecting two fields from the description, which may be potentially related, as an independent variable and a response variable of the models, and instructions for converting categorical fields in the description into factors of the models.

43. The computer system of claim 42, wherein the set of predefined heuristics further comprises:

instructions for defining a first quantitative field on the row (or the column) as a response variable and a second field on the column (or the row) as an independent variable if the first quantitative field is the only quantitative field on the row or the column of the view and the second field is either a quantitative field or a categorical field having a quantitative interpretation; and

instructions for generating the one or more models using data in the dataset associated with the response variable and the independent variable.

44. The computer system of claim 42, wherein the set of predefined heuristics further comprises:

instructions for defining a first quantitative field on the row (or the column) as a response variable and a second quantitative field on the column (or the row) as an independent variable if the second quantitative field is a quantitative field from a dimension class; and

instructions for generating the one or more models using data in the dataset associated with the response variable and the independent variable.

45. The computer system of claim 42, wherein the set of predefined heuristics further comprises:

instructions for defining a first quantitative field on the row as a response variable and a second quantitative field on the column as an independent variable; and

instructions for generating the one or more models using data in the dataset associated with the response variable and the independent variable.

46. The computer system of claim 45, wherein the set of predefined heuristics further comprises:

instructions for repeating said defining and generating instructions for each unique pair of first and second quantitative fields on the column and row.

47. The computer system of claim 42, wherein the set of predefined heuristics further comprises:

instructions for defining each categorical field on the row, the column and the page of the view as a categorical factor of the one or more models.

48. The computer system of claim 42, wherein the set of predefined heuristics further comprises:

instructions for defining each categorical field in the encoding area as a categorical factor of the one or more models if there are more than one data points corresponding to each value in the categorical field.

49. The computer system of claim 41, further comprising:

instructions for ignoring each quantitative field in the encoding area of the view when translating the description into the one or more models.

50. The computer system of claim 39, wherein each of the one or more models has a model type including linear, polynomial, exponential, logarithmic, and transcendental.

51. The computer system of claim 39, further comprising:

instructions for displaying the view of the dataset in accordance with the description; and

instructions for displaying graphs corresponding to the one or more models.

52. The computer system of claim 39, further comprising:

instructions for determining if it is possible to translate the description into any model by examining the properties of the fields associated with the view.

53. The computer system of claim 39, wherein the one or more models are statistical models.

54. The computer system of claim 39, further comprising:

instructions for generating a plurality of models using the description, each model having a specific model type and a set of parameters; and

instructions for automatically selecting from the plurality of models one having a simpler model type and fewer parameters according to predefined criteria.

55. The computer system of claim 39, wherein the description is generated by a user dragging and dropping a subset of the fields into respective field containers of a graphical user interface.

56. The computer system of claim 39, wherein the description is generated by a user selecting a subset of the fields as fields of interest through a graphical user interface.

57. The computer system of claim 39, wherein the description is generated by a user entering a text string including a subset of the fields as fields of interest through a user interface.

58. A computer program product for use in conjunction with a computer system, the computer program product comprising a computer readable storage medium and a computer program mechanism embedded therein, the computer program mechanism comprising:

instructions for generating a description to visualize a dataset, the description including multiple fields associated with a view of the dataset;

instructions for determining a set of properties for each of the multiple fields; and

instructions for automatically translating the description into one or more models based on the respective properties of the multiple fields and a set of predefined heuristics.

59. The computer program product of claim 58, wherein the multiple fields are partitioned into a plurality of sets, each set associated with a respective region of the view including row, column, page, and an encoding area.

60. The computer program product of claim 59, wherein each field includes a field type selected from the group consisting of categorical and quantitative and a data type selected from the group consisting of integer, real, Boolean, and time.

61. The computer program product of claim 60, wherein the set of predefined heuristics includes instructions for selecting two fields from the description, which may be potentially related, as an independent variable and a response variable of the models, and instructions for converting categorical fields in the description into factors of the models.

62. The computer program product of claim 61, wherein the set of predefined heuristics further comprises:

instructions for defining a first quantitative field on the row (or the column) as a response variable and a second field on the column (or the row) as an independent variable if the first quantitative field is the only quantitative field on the row or the column of the view and the second field is either a quantitative field or a categorical field having a quantitative interpretation; and

instructions for generating the one or more models using data in the dataset associated with the response variable and the independent variable.

63. The computer program product of claim 61, wherein the set of predefined heuristics further comprises:

instructions for defining a first quantitative field on the row (or the column) as a response variable and a second quantitative field on the column (or the row) as an independent variable if the second quantitative field is a quantitative field from a dimension class; and

instructions for generating the one or more models using data in the dataset associated with the response variable and the independent variable.

64. The computer program product of claim 61, wherein the set of predefined heuristics further comprises:

instructions for defining a first quantitative field on the row as a response variable and a second quantitative field on the column as an independent variable; and

instructions for generating the one or more models using data in the dataset associated with the response variable and the independent variable.

65. The computer program product of claim 64, wherein the set of predefined heuristics further comprises:

instructions for repeating said defining and generating instructions for each unique pair of first and second quantitative fields on the column and row.

66. The computer program product of claim 61, wherein the set of predefined heuristics further comprises:

instructions for defining each categorical field on the row, the column and the page of the view as a categorical factor of the one or more models.

67. The computer program product of claim 61, wherein the set of predefined heuristics further comprises:

instructions for defining each categorical field in the encoding area as a categorical factor of the one or more models if there are more than one data points corresponding to each value in the categorical field.

68. The computer program product of claim 61, wherein the set of predefined heuristics further comprises:

instructions for ignoring each quantitative field in the encoding area of the view when translating the description into the one or more models.

69. The computer program product of claim 58, wherein each of the one or more models has a model type including linear, polynomial, exponential, logarithmic, and transcendental.

70. The computer program product of claim 58, further comprising:

instructions for displaying the view of the dataset in accordance with the description; and

instructions for displaying graphs corresponding to the one or more models.

71. The computer program product of claim 58, further comprising:

instructions for determining if it is possible to translate the description into any model by examining the properties of the fields associated with the view.

72. The computer program product of claim 58, wherein the one or more models are statistical models.

73. The computer program product of claim 58, further comprising:

instructions for generating a plurality of models using the description, each model having a specific model type and a set of parameters; and

instructions for automatically selecting from the plurality of models one having a simpler model type and fewer parameters according to predefined criteria.

74. The computer program product of claim 58, wherein the description is generated by a user dragging and dropping a subset of the fields into respective field containers of a graphical user interface.

75. The computer program product of claim 58, wherein the description is generated by a user selecting a subset of the fields as fields of interest through a graphical user interface.

76. The computer program product of claim 58, wherein the description is generated by a user entering a text string including a subset of the fields as fields of interest through a user interface.

77. A computer system comprising:

means for providing a description of a view of a dataset, the description including multiple fields associated with the dataset;

means for determining a set of properties for each of the multiple fields; and

means for automatically translating the description into one or more models based on the respective properties of the multiple fields and a set of predefined heuristics.