Granted Patent B2

US 9,779,524 · App. 13/745,985 · Granted Oct 3, 2017

Abstract

Claims (46)

1. A method comprising:

determining, by a system having a processor, a number of events that share a first collection of values of plural attributes;

computing, by the system using a contingency calculation based on the number of events, a discriminative metric representing a statistical significance of the events that share the first collection of values of the plural attributes, the statistical significance represented by the computed discriminative metric indicating that the events that share the first collection of values of the plural attributes have a higher expected occurrence than events sharing a different collection of values of the plural attributes;

displaying, by the system, a visualization that includes cells representing respective events, the visualization including a region containing a subset of the cells, and the visualization including plural axes respectively representing the plural attributes, wherein cells of the subset represent the events that share the first collection of values of the plural attributes;

assigning, by the system, a visual characteristic of a significance visual indicator based on a value of the computed discriminative metric, the significance visual indicator placed in the visualization and associated with the region to indicate the statistical significance of the events sharing the first collection of values of the plural attributes; and

responsive to user selection of the significance visual indicator in the visualization, causing display, by the system, of information regarding the subset of the cells associated with the selected significance visual indicator.

2. The method of claim 1 , further comprising:

creating a contingency data structure having a first entry containing the number of events that share the collection of values of the plural attributes,

wherein computing the discriminative metric uses the contingency data structure.

3. The method of claim 2 , wherein the first collection of values includes a first value of a first attribute represented by a first axis of the visualization, and a second value of a second attribute represented by a second axis of the visualization, and wherein the contingency data structure further includes additional entries that include:

a second entry containing a number of events that share the first value of the first attribute and other values of the second attribute,

a third entry containing a number of events that share the second value of the second attribute and other values of the first attribute, and

a fourth entry containing a number of events that share the other values of the first attribute and the other values of the second attribute.

4. The method of claim 1 , wherein computing the discriminative metric comprises computing an odds ratio metric that is computed from a ratio of a first value to a second value, the first value based on the number of events that share the first collection of values of the plural attributes, and the second value based on a number of events that share the different collection of values of the plural attributes.

5. The method of claim 1 , wherein the significance visual indicator includes a graphical element having a brightness, the method comprising:

setting a degree of the brightness based on the value of the discriminative metric.

6. The method of claim 5 , wherein the graphical element includes a ring around the region containing the subset of cells, the ring having the set degree of the brightness.

7. The method of claim 1 , further comprising:

assigning different colors to the cells based on corresponding values of a further attribute of the events.

8. The method of claim 7 , further comprising:

sorting the cells in the subset according to respective values of the further attribute, wherein the cells in the region are placed based on the sorting, and wherein placing the cells in the region based on the sorting causes sub-groups of the cells to be formed, wherein each of the sub-groups correspond to cells sharing a common value of the further attribute.

9. An article comprising at least one non-transitory machine-readable storage medium storing instructions that upon execution cause a system to:

create a data structure having parameters representing respective numbers of occurrences of collections of values of plural attributes occurring in events;

compute, using a contingency calculation based on the data structure, a discriminative metric representing a statistical significance of events that share a first collection of the collections of values of the plural attributes;

generate a visualization that includes cells representing respective events, the visualization including a region containing a subset of the cells, and the visualization including plural axes respectively representing the plural attributes, wherein cells of the subset represent the events that share the first collection of values of the plural attributes;

assign a visual characteristic of a significance visual indicator based on a value of the computed discriminative metric, the significance visual indicator placed in the visualization and associated with the region to indicate the statistical significance, as represented by the computed discriminative metric, of the events that share the first collection of values of the plural attributes; and

responsive to user selection of the significance visual indicator in the visualization, cause display of information regarding the events represented by the subset of the cells associated with the selected significance visual indicator.

10. The article of claim 9 , wherein the visualization includes regions for respective groups of cells that correspond to respective collections of values of the plural attributes, and wherein the instructions upon execution cause the system to further decide not to provide a significance visual indicator for at least one of the regions.

11. The article of claim 10 , wherein deciding to not provide a significance visual indicator for the at least one region is based on determining that a value of the discriminative metric representing a statistical significance of events corresponding to the at least one region does not satisfy a specified criterion.

12. The article of claim 10 , wherein a size of each of the regions is indicative of a corresponding frequency of occurrence of events sharing the respective collection of attribute values, and wherein the statistical visual indicator is a ring around a first region of the regions, the statistical visual indicator distinct from the size of the first region.

13. The article of claim 9 , wherein generating the visualization further comprises assigning colors to the cells depending upon respective values of a further attribute.

14. The article of claim 9 , wherein the significance visual indicator has a brightness that depends upon the value of the computed discriminative metric.

15. A system comprising:

at least one processor; and

a non-transitory storage medium storing instructions executable on the at least one processor to:

determine a quantity of events that share a first collection of values of plural attributes;

compute, based on a contingency data structure, a discriminative metric representing a statistical significance of the events that share the first collection of values of the plural attributes, wherein the contingency data structure includes an entry containing the quantity of events that share the first collection of values of the plural attributes, the statistical significance represented by the computed discriminative metric indicating that the events that share the first collection of values of the plural attributes have a higher expected occurrence than events sharing a different collection of values of the plural attributes;

cause display of a visualization that includes cells representing respective events, the visualization including a region containing a subset of the cells, and the visualization including plural axes respectively representing the plural attributes, wherein cells of the subset represent the events that share the first collection of values of the plural attributes;

assign a visual characteristic of a significance visual indicator based on a value of the computed discriminative metric, the significance visual indicator included in the visualization and associated with the region to indicate the statistical significance of the events sharing the first collection of values of the plural attributes; and

responsive to user selection of the significance visual indicator in the visualization, cause display of information regarding the subset of the cells associated with the selected significance visual indicator.

16. The system of claim 15 , wherein the significance visual indicator includes a graphical element having a brightness, wherein the instructions are executable on the at least one processor to further set a degree of the brightness based on the value of the discriminative metric, wherein the brightness of the graphical element varies according to different values of the discriminative metric.

17. The system of claim 16 , wherein the graphical element includes a ring around the region containing the subset of the cells.

18. The system of claim 15 , wherein the instructions are executable on the at least one processor to further:

assign different colors to the cells based on corresponding values of a further attribute of the events.

19. The system of claim 15 , wherein the computing of the discriminative metric comprises computing an odds ratio metric that is computed from a ratio of a first value to a second value, the first value based on the quantity of events that share the first collection of values of the plural attributes, and the second value based on a quantity of events that share the different collection of values of the plural attributes,

the data structure including another entry that includes the quantity of events that share the different collection of values of the plural attributes.

Assignments (13)