Method and system for dynamic visualization of multi-dimensional data

• US 7,265,755 B2
• Filed: 12/17/2003
• Issued: 09/04/2007
• Est. Priority Date: 12/17/2002
• Status: Expired due to Fees

• Current Assignee: Terastat Incorporated
• Original Assignee: Terastat Incorporated
• Inventors: Peterson, Bruce A.
• Primary Examiner(s): Tung; Kee M.
• Assistant Examiner(s): Amini; J.
• Application Number:
  US10/739,897
• Publication Number:
  US 20040183800A1
• Time in Patent Office: 1,357 Days
• Field of Search: 345/473, 345/440, 715/700
• US Class Current: 345/440
• CPC Class Codes: G06T 11/206 Drawing of charts or graphs

Abstract

Methods and systems for dynamic visualization of highly dimensional data are provided. Example embodiments provide a Dynamic Visualization System (a “DVS”), which maps data attributes to visualization dimensions and automatically imputes attribute values on-the-fly for each value of a sequencing dimension. Data having different resolutions can be simultaneously displayed in the same visualization. In one embodiment, the DVS comprises a Dynamic Data Management System with a user interface for specifying and mapping attributes, and one or more Dynamic Visualization Engines with near real-time control user interfaces. These components cooperate to define visualization scenarios based upon specified data subsets, determine and automatically execute a defined visualization, and allow analysts and other users to interactively modify the visualization.

Claims

1. A method in a computer system for automatically presenting a dynamic visualization of data in a multi-dimensional space of greater than three dimensions, the data having a plurality of attributes, comprising:

receiving a plurality of mappings of data attributes to visualization dimensions, wherein the visualization dimensions include at least one appearance dimension, a plurality of spatial dimensions, and at least one sequencing dimension;

determining a plurality of data entities from the data, each data entity associated with a portion of the data; and

for each determined data entity,

at a time of visualization of the data, automatically and dynamically generating a series of representations of the data entity in the multi-dimensional space and automatically and dynamically presenting the generated series of representations, the representations based upon values of each of the data attributes of the data associated with the data entity that have been mapped to the appearance, spatial, and sequencing dimensions, to portray changes in the data entity over values of the sequencing dimension so that trends in the data can be identified.

2. The method of claim 1 wherein the series of representations includes an indication of a history of changes to the data entity over previous values of the sequencing dimension.

3. The method of claim 2 wherein the history of changes is indicated by a trail from a prior representation of the data entity that corresponds to at least one earlier value of the sequencing dimension to a current value of the sequencing dimension.

4. The method of claim 3 wherein the trail shows movement in at least one spatial dimension.

5. The method of claim 3 wherein the trail shows changes in at least one appearance dimension.

6. The method of claim 3 wherein the length of the trail is proportional to the magnitude of change of the data entity over a period defined by values of the sequencing dimension.

7. The method of claim 1 wherein the sequencing dimension is time and wherein the automatically and dynamically generating the series of representations of the data entity in the multi-dimensional space and automatically and dynamically presenting the generated series of representations portrays changes in the data entity over values of time.

8. The method of claim 1 wherein the automatically and dynamically generating the series of representations of the data entity in the multi-dimensional space, further comprises:

for each generated representation of the data entity, automatically and dynamically imputing attribute values, based upon a current value of the sequencing dimension, for each attribute of the data associated with the data entity that is mapped to a spatial dimension and for each attribute of the data associated with the data entity that is mapped to an appearance dimension.

9. The method of claim 8 wherein the automatically and dynamically imputing attribute values comprises automatically and dynamically imputing values based upon at least one imputation function derived from the data associated with the data entity.

10. The method of claim 9 wherein the imputation function is an interpolation function.

11. The method of claim 9 wherein the imputation function is at least one of a univariate function of data entity data, a multivariate function of data from a plurality of data entities, and a function that preserves statistical characteristics of data entity data.

12. The method of claim 9 wherein the imputation function imputes a value by examining a plurality of existing data sequences.

13. The method of claim 9 wherein the automatically and dynamically imputing attribute values comprises, for each attribute of the data associated with the data entity that is mapped to at least one of a spatial and an appearance dimension, automatically and dynamically imputing a value based upon a function derived from the data associated with that attribute, wherein at least two attributes impute values based upon different functions.

14. The method of claim 1 wherein the automatically and dynamically generating the series of representations of the data entity in the multi-dimensional space and automatically and dynamically presenting the generated series of representations, further comprises:

instantiating a data object that corresponds to each data entity, each object having a plurality of instantiated spatial attribute and appearance attribute objects, each attribute object having an instantiated imputation object that computes a value for a visualization dimension dynamically;

determining a next sequencing dimension value; and

for each instantiated data object,

causing each of the instantiated imputation objects to determine an attribute value that corresponds to the next sequencing dimension value; and

causing the data entity to be presented according to the determined attribute values of the corresponding instantiated data object.

15. The method of claim 14, further comprising:

repeating the steps of determining the next sequencing dimension value and for each instantiated object, causing each of the instantiated imputation objects to determine the corresponding attribute value and causing the data entity to be presented.

16. The method of claim 1, further comprising:

determining and presenting a visual filter that defines criteria of interest; and

indicating when the attribute values of a data entity that are mapped to a visualization dimension intersect with the criteria defined by the visual filter.

17. The method of claim 16, the indicating when the attribute values intersect with the criteria defined by the visual filter, further comprising:

indicating at least one of that the attribute values intersect with the criteria defined by the visual filter and that the attributes cease to intersect with the criteria defined by the visual filter by triggering an event.

18. The method of claim 17 wherein the event comprises storing an indication of the data entity whose attribute values intersect with the criteria.

19. The method of claim 17 wherein the event comprises at least one of a notification, logged data, and execution of code.

20. The method of claim 16, the determining and presenting the visual filter that defines criteria of interest, further comprising:

determining and presenting a visual filter that defines at least one region of spatial dimension values.

21. The method of claim 16, the determining and presenting the visual filter that defines criteria of interest, further comprising:

determining and presenting a visual filter that defines at least one set of appearance dimension values.

22. The method of claim 21 wherein the appearance dimension values comprise at least one of an indication of size, color, glyph, pattern, texture, and shape.

23. The method of claim 1, the determining of the plurality of data entities further comprising:

receiving an indication of a data attribute of the data that is to be used to select a plurality of data entities; and

determining a plurality of data entities by associating a unique data entity with each unique value of the indicated data attribute.

24. The method of claim 1, the visualization dimensions further comprising at least one audio dimension, wherein at least one attribute is mapped to the audio dimension, and wherein the automatically and dynamically generating the series of representations of the entity in the multi-dimensional space and automatically and dynamically presenting the generated series of representations, further comprises:

automatically and dynamically generating a series of representations of the data entity in the multi-dimensional space and automatically and dynamically presenting the generated series of representations, the representations based upon values of each of the data attributes of the data associated with the data entity that have been mapped to the appearance, spatial, sequencing, and audio dimensions.

25. The method of claim 1, wherein the appearance dimension is at least one of size, color, glyph, pattern, shape, and texture.

26. The method of claim 1, each spatial and appearance dimension having at least one display parameter, further comprising:

while the series of representations of the data entity are being presented,

receiving an indication of a change to the at least one display parameter of at least one of the spatial dimensions and the appearance dimension; and

adjusting the presentation of the series of representations of the data entity to account for the indicated change.

27. A computer-readable memory medium containing instructions that, when executed, control a computer processor to automatically present a dynamic visualization of data in a multi-dimensional space of greater than three dimensions, the data having a plurality of attributes, by performing a method comprising:

receiving a plurality of mappings of data attributes to visualization dimensions, wherein the visualization dimensions include at least one appearance dimension, a plurality of spatial dimensions, and at least one sequencing dimension;

determining a plurality of data entities from the data, each data entity associated with a portion of the data; and

for each determined data entity,

automatically and dynamically generating a series of representations of the data entity in the multi-dimensional space and automatically and dynamically presenting the generated series of representations, the representations based upon values of each of the data attributes of the data associated with the data entity that have been mapped to the appearance, spatial, and sequencing dimensions, to portray changes to the data entity over values of the sequencing dimension so that trends in the data can be identified.

28. The computer-readable memory medium of claim 27 wherein the series of representations includes an indication of a history of changes to the data entity over previous values of the sequencing dimension.

29. The computer-readable memory medium of claim 28 wherein the history of changes is indicated by a trail from a prior representation of the data entity that corresponds to at least one earlier value of the sequencing dimension to a current value of the sequencing dimension.

30. The computer-readable memory medium of claim 29 wherein the trail shows movement in at least one spatial dimension.

31. The computer-readable memory medium of claim 29 wherein the trail shows changes in at least one appearance dimension.

32. The computer-readable memory medium of claim 29 wherein the length of the trail is proportional to the magnitude of change of the data entity over a period defined by values of the sequencing dimension.

33. The computer-readable memory medium of claim 27 wherein the sequencing dimension is time and wherein the automatically and dynamically generating the series of representations of the data entity in the multi-dimensional space and automatically and dynamically presenting the generated series of representations portray changes in the data entity over values of time.

34. The computer-readable memory medium of claim 27 wherein the automatically and dynamically generating the series of representations of the data entity in the multi-dimensional space further comprises:

for each generated representation of the data entity, automatically and dynamically imputing attribute values, based upon a current value of the sequencing dimension, for each attribute of the data associated with the data entity that is mapped to a spatial dimension and for each attribute of the data associated with the data entity that is mapped to an appearance dimension.

35. The computer-readable memory medium of claim 34 wherein the automatically and dynamically imputing attribute values further comprise automatically and dynamically imputing values based upon at least one imputation function derived from the data associated with the data entity.

36. The computer-readable memory medium of claim 35 wherein the imputation function is an interpolation function.

37. The computer-readable memory medium of claim 35 wherein the imputation function is at least one of a univariate function of data entity data, a multivariate function of data from a plurality of data entities, and a function that preserves statistical characteristics of data entity data.

38. The computer-readable memory medium of claim 35 wherein the imputation function imputes a value by examining a plurality of existing data sequences.

39. The computer-readable memory medium of claim 34 wherein the automatically and dynamically imputing attribute values further comprises, for each attribute of the data associated with the data entity that is mapped to at least one of a spatial and an appearance dimension, automatically and dynamically imputing a value based upon a function derived from the data associated with that attribute, wherein at least two attributes impute values based upon different functions.

40. The computer-readable memory medium of claim 27 wherein the automatically and dynamically generating the series of representations of the data entity in the multi-dimensional space and automatically and dynamically presenting the generated series of representations, further comprises:

instantiating a data object that corresponds to each data entity, each object having a plurality of instantiated spatial attribute and appearance attribute objects, each attribute object having an instantiated imputation object that computes a value for a visualization dimension dynamically;

determining a next sequencing dimension value; and

for each instantiated data object,

causing each of the instantiated imputation objects to determine an attribute value that corresponds to the next sequencing dimension value; and

causing the data entity to be presented according to the determined attribute values of the corresponding instantiated data object.

41. The computer-readable memory medium of claim 40 wherein the method further comprises:

repeating the steps of determining the next sequencing dimension value and for each instantiated object, causing each of the instantiated imputation objects to determine the corresponding attribute value and causing the data entity to be presented.

42. The computer-readable memory medium of claim 27 wherein the method further comprises:

determining and presenting a visual filter that defines criteria of interest; and

indicating when the attribute values of a data entity that are mapped to a visualization dimension intersect with the criteria defined by the visual filter.

43. The computer-readable memory medium of claim 42, the indicating when the attribute values intersect with the criteria defined by the visual filter further comprising:

indicating at least one of that the attribute values intersect with the criteria defined by the visual filter and that the attribute values cease to intersect with the criteria defined by the visual filter by triggering an event.

44. The computer-readable memory medium of claim 43 wherein the event comprises storing an indication of the data entity whose attribute values intersect with the criteria.

45. The computer-readable memory medium of claim 43 wherein the event comprises at least one of a notification, logged data, and execution of code.

46. The computer-readable memory medium of claim 42, the determining and presenting the visual filter that defines criteria of interest;

further comprising;

determining and presenting a visual filter that defines at least one region of spatial dimension values.

47. The computer-readable memory medium of claim 42, the determining and presenting the visual filter that defines criteria of interest further comprising:

determining and presenting a visual filter that defines at least one set of appearance dimension values.

48. The computer-readable memory medium of claim 47 wherein the appearance dimension values comprise at least one of an indication of size, color, glyph, pattern, texture, and shape.

49. The computer-readable memory medium of claim 27, the determining of the plurality of data entities further comprising:

receiving an indication of a data attribute of the data that is to be used to select a plurality of data entities; and

determining a plurality of data entities by associating a unique data entity with each unique value of the indicated data attribute.

50. The computer-readable memory medium of claim 27, the visualization dimensions further comprising at least one audio dimension, wherein at least one attribute is mapped to the audio dimension, and wherein the automatically and dynamically generating the series of representations of the entity in the multi-dimensional space and automatically and dynamically presenting the generated series of representations further comprise:

automatically and dynamically generating a series of representations of the data entity in the multi-dimensional space and automatically and dynamically presenting the generated series of representations, the representations based upon values of each of the data attributes of the data associated with the data entity that have been mapped to the appearance, spatial, sequencing, and audio dimensions.

51. The computer-readable memory medium of claim 27 wherein the appearance dimension is at least one of size, color, glyph, pattern, shape, and texture.

52. The computer-readable memory medium of claim 27, each spatial and appearance dimension having at least one display parameter, the method further comprising:

while the series of representations of the data entity are being presented,

receiving an indication of a change to the at least one display parameter of at least one of the spatial dimensions and the appearance dimension; and

adjusting the presentation of the series of representations of the data entity to account for the indicated change.

53. A data visualization system for automatically presenting a dynamic visualization of data in a multi-dimensional space of greater than three dimensions, the data having a plurality of data attributes and stored in a data repository, comprising:

an attribute mapping mechanism that is structured to receive a plurality of mappings of data attributes to visualization dimensions, wherein the visualization dimensions include at least one appearance dimension, a plurality of spatial dimensions, and at least one sequencing dimension;

an entity management component that is structured to determine from the data stored in the data repository a plurality of data entities, each data entity associated with a portion of the data, and that instantiates a data object that corresponds to each determined data entity and associated data; and

a visualization execution mechanism that is structured to, for each value of the at least one sequencing dimension, automatically invoke each instantiated data object to automatically and dynamically determine and present a representation of the corresponding data entity in the multi-dimensional space based upon values of each of the data attributes of the data associated with the instantiated data object that have been mapped to the appearance, spatial, and sequencing dimensions.

54. The data visualization system of claim 53 wherein the representation presented by each instantiated data object includes an indication of a history of changes to the data entity over previous values of the sequencing dimension.

55. The data visualization system of claim 54 wherein the history of changes is indicated by a trail from a prior representation of the data entity that corresponds to at least one earlier value of the sequencing dimension to a current value of the sequencing dimension.

56. The data visualization system of claim 55 wherein the trail shows movement in at least one spatial dimension.

57. The data visualization system of claim 55 wherein the trail shows changes in at least one appearance dimension.

58. The data visualization system of claim 55 wherein the length of the trail is proportional to the magnitude of change of the data entity over a period defined by values of the sequencing dimension.

59. The data visualization system of claim 53 wherein the sequencing dimension is time and wherein the representation presented by each instantiated object portrays changes in the corresponding data entity over values of time.

60. The data visualization system of claim 53 wherein each instantiated object automatically and dynamically determines and presents a representation of the corresponding data entity by automatically and dynamically imputing attribute values, based upon a current value of the sequencing dimension, for each attribute of the associated data that is mapped to a spatial dimension and for each attribute of the associated data that is mapped to an appearance dimension.

61. The data visualization system of claim 60 wherein the automatically and dynamically imputing attribute values comprises automatically and dynamically imputing values based upon at least one imputation function derived from the associated data.

62. The data visualization system of claim 61 wherein the imputation function is an interpolation function.

63. The data visualization system of claim 61 wherein the imputation function is at least one of a univariate function of data entity data, a multivariate function of data from a plurality of data entities, and a function that preserves statistical characteristics of data entity data.

64. The data visualization system of claim 61 wherein the imputation function imputes a value by examining a plurality of existing data sequences.

65. The data visualization system of claim 60 wherein the automatically and dynamically imputing attribute values comprises, for each attribute of the associated data that is mapped to at least one of a spatial and an appearance dimension, automatically and dynamically imputing a value based upon a function derived from the data associated with that attribute, wherein at least two attributes impute values based upon different functions.

66. The data visualization system of claim 53, wherein the automatically and dynamically imputing attribute values for each instantiated data object is performed by invoking a plurality of imputation objects that correspond to the instantiated data object, each imputation object corresponding to a data attribute of the data associated with the data object that has been mapped to a visualization dimension.

67. The data visualization system of claim 53, further comprising:

a visual filter that is structured to define criteria of interest and to indicate when the attribute values of a data entity that are mapped to a visualization dimension at least one of intersect and cease to intersect with the criteria defined by the visual filter.

68. The data visualization system of claim 67 wherein the visual filter indicates when the attribute values intersect with the criteria defined by the visual filter by triggering an event.

69. The data visualization system of claim 68 wherein the event comprises storing an indication of the data entity whose attribute values intersect with the criteria.

70. The data visualization system of claim 68 wherein the event is at least one of a notification, logged data.

71. The data visualization system of claim 67 wherein the visual filter defines criteria of interest by defining at least one region of spatial dimension values.

72. The data visualization system of claim 67 wherein the visual filter defines criteria of interest by defining at least one set of appearance dimension values.

73. The data visualization system of claim 72 wherein the appearance dimension values comprise at least one of an indication of size, color, glyph, pattern, texture, and shape.

74. The data visualization system of claim 53 wherein the entity management component determines the plurality of data entities by receiving an indication of a data attribute of the data that is to be used to select a plurality of data entities and associating a unique data entity with each unique value of the indicated data attribute.

75. The data visualization system of claim 53, the visualization dimensions further comprising at least one audio dimension, wherein the attribute mapping mechanism receives a mapping of at least one attribute to the audio dimension, and wherein the representations of the data entity presented by each instantiated data object are also based upon values of the data attribute of the associated data of each data object that has been mapped to the audio dimension.

76. The data visualization system of claim 53, wherein the appearance dimension is at least one of size, color, glyph, pattern, shape, and texture.

77. The data visualization system of claim 53, each spatial and appearance dimension having at least one display parameter, wherein the visualization execution mechanism is further structured to, while the series of representations of each data entity are being presented by each data object,

receive an indication of a change to the at least one display parameter of at least one of the spatial dimensions and the appearance dimension;

and

forward the indicated change to each instantiated data object so each data object accordingly adjusts the representations of the data entity to account for the indicated change when presented.

78. A method in a computer system for visually indicating data of interest from among multi-dimensional data, comprising:

displaying a dynamic visualization of the multi-dimensional data;

defining a visual filter that defines criteria of interest by specifying a range of values for at least one of an appearance dimensions and a spatial dimension;

displaying delineations of at least one convex region, the displayed delineations indicating the defined criteria of interest; and

upon receiving an indication that values of the multi-dimensional data displayed in the dynamic visualization intersect with the defined criteria of interest, displaying a representation of the multi-dimensional data having the indicated values within the displayed delineated region.

79. The method of claim 78 wherein the delineated region is a convex geometric shape.

80. The method of claim 79 wherein the convex geometric shape is at least one of an ellipsoid, a sphere, and a rectangle.

81. The method of claim 78 wherein the display of the delineations displays at least one of a frame and an outline.

82. The method of claim 78, further comprising:

receiving an indication that the values of the multi-dimensional data displayed in the dynamic visualization cease to intersect with the defined criteria of interest; and

triggering an event.

83. The method of claim 78 wherein the criteria of interest is a set of values of an appearance dimension.

84. The method of claim 83 wherein the appearance dimension is at least one of a size, color, pattern, texture, shape, and glyph.

85. The method of claim 78 wherein the criteria of interest defines sets of ranges for each of three spatial dimensions.

86. The method of claim 78, further comprising:

upon receiving the indication that values of the multi-dimensional data displayed in the dynamic visualization intersect with the defined criteria of interest, triggering an event.

87. The method of claim 86 wherein the event is at least one of a notification, logged data, and execution of code.

88. The method of claim 86 wherein the event is storing an indication of the data whose values intersect with the defined criteria of interest.

89. The method of claim 78 wherein the displayed dynamic visualization of the multi-dimensional data is an animated display of multi-dimensional data over a sequence of values.

90. The method of claim 78 wherein the displayed dynamic visualization of the multi-dimensional data is portrayal of multi-dimensional data over a sequence of values.

91. A computer-readable memory medium containing instructions that, when executed, control a computer processor to visually indicate data of interest from among displayed multi-dimensional data, by performing a method comprising:

displaying a dynamic visualization of the multi-dimensional data;

defining a visual filter that defines criteria of interest by specifying a range of values for at least one of an appearance dimensions and a spatial dimension;

displaying delineations of at least one convex region, the displayed delineations indicating the defined criteria of interest; and

upon receiving an indication that values of the multi-dimensional data displayed in the dynamic visualization intersect with the defined criteria of interest, displaying a representation of the multi-dimensional data having the indicated values within the displayed delineated region.

92. The computer-readable memory medium of claim 91 wherein the delineated region is a convex geometric shape.

93. The computer-readable memory medium of claim 92 wherein the convex geometric shape is at least one of an ellipsoid, a sphere, or a rectangle.

94. The computer-readable memory medium of claim 91 wherein the display of the delineations displays at least one of a frame or an outline.

95. The computer-readable memory medium of claim 91, the method further comprising:

receiving an indication that the values of the multi-dimensional data displayed in the dynamic visualization cease to intersect with the defined criteria of interest; and

triggering an event.

96. The computer-readable memory medium of claim 91 wherein the criteria of interest is a set of values of an appearance dimension.

97. The computer-readable memory medium of claim 96 wherein the appearance dimension is at least one of a size, color, pattern, texture, shape, or glyph.

98. The computer-readable memory medium of claim 91 wherein the criteria of interest defines sets of ranges for each of three spatial dimensions.

99. The computer-readable memory medium of claim 91, the method further comprising:

upon receiving the indication that values of the multi-dimensional data displayed in the dynamic visualization intersect with the defined criteria of interest, triggering an event.

100. The computer-readable memory medium of claim 99 wherein the event is at least one of a notification, logged data or execution of code.

101. The computer-readable memory medium of claim 99 wherein the event is storing an indication of the data whose values intersect with the defined criteria of interest.

102. The computer-readable memory medium of claim 91 wherein the multi-dimensional data displayed in the dynamic visualization is an animated display of multi-dimensional data over a sequence of values.

103. The computer-readable memory medium of claim 91 wherein the multi-dimensional data displayed in the dynamic visualization is a portrayal of multi-dimensional data over a sequence of values.

104. A computer-based visual filter for visually indicating data of interest from among multi-dimensional data, comprising:

input mechanism that defines criteria of interest by receiving an indication of a range of values for at least one of an appearance dimensions and a spatial dimension;

a display mechanism that is structured to,

display a dynamic visualization of the multi-dimensional data;

display a convex delineation, the displayed delineation indicating the defined criteria of interest; and

upon receiving an indication that values of the multi-dimensional data displayed in the dynamic visualization intersect with the defined criteria of interest, display a representation of the multi-dimensional data having the indicated values within the displayed delineation.

105. The visual filter of claim 104 wherein the delineation is a convex geometric shape.

106. The visual filter of claim 105 wherein the convex geometric shape is at least one of an ellipsoid, a sphere or a rectangle.

107. The visual filter of claim 104 wherein the display of the geometric shape displays at least one of a frame or an outline.

108. The visual filter of claim 104 wherein the display mechanism is further structured to indicate that values of the displayed multi-dimensional data cease to intersect with the defined criteria of interest.

109. The visual filter of claim 108 wherein the indication is a triggered event.

110. The visual filter of claim 104 wherein the criteria of interest is a set of values of an appearance dimension.

111. The visual filter of claim 110 wherein the appearance dimension is at least one of a size, color, pattern, texture, shape, or glyph.

112. The visual filter of claim 104 wherein the criteria of interest defines sets of ranges for each of three spatial dimensions.

113. The visual filter of claim 104, further comprising:

an alert mechanism that, upon receiving the indication that values of the multi-dimensional data displayed in the dynamic visualization intersect with the defined criteria of interest, triggers an event.

114. The visual filter of claim 113 wherein the event is at least one of a notification, logged data, and execution of code.

115. The visual filter of claim 113 wherein the event is storing an indication of the data whose values intersect with the defined criteria of interest.

116. The visual filter of claim 104 wherein the multi-dimensional data displayed in the dynamic visualization is an animated display of multi-dimensional data over a sequence of values.

117. The visual filter of claim 104 wherein the multi-dimensional data displayed in the dynamic visualization is a portrayal of multi-dimensional data over a sequence of values.