Dynamically generating visualizations of data based on correlation measures and search history
Described embodiments provide systems and methods for generating visualizations of data based on correlation measures and search history. An analysis engine may access data observed from a data source over time. The analysis engine may determine a variation of each of at least a first metric and a second metric of the data, over time. A correlation engine may determine a correlation measure between the first metric and a second metric, over time. The correlation engine may generate, responsive to the correlation measure being greater than a reference level, a visualization of the first metric and the second metric varying in time, on a device to display to a user.
1. A method comprising:
accessing, by at least one processor, data observed from a data source over time;
determining, by the at least one processor, that a variation of at least a first metric or a second metric of the data over time is greater than a threshold;
determining, by the at least one processor responsive to determining that the variation is greater than the threshold, that a correlation measure between the first metric and the second metric over time is greater than a reference level;
selecting, by the at least one processor responsive to determining that the correlation measure is greater than the reference level, the first metric and the second metric for visualization;
generating, by the at least one processor, responsive to selection, a visualization of the first metric and the second metric varying in time, on a device to display to a user;
determining, by the at least one processor, a level of interest for at least one of the first metric or the second metric according to a number of user interactions with the visualization;
modifying, by the at least one processor, the correlation measure between the first metric and the second metric based at least on the level of interest; and
updating, by the at least one processor, the visualization in accordance with the modified correlation measure.
2. The method of claim 1 , further comprising:
detecting, by the at least one processor, that the data source comprises a new data source; and
dynamically generating, by the at least one processor responsive to the new data source, the visualization of the first metric and the second metric varying in time, to display to the user.
3. The method of claim 1 , comprising accessing the data observed from the data source over time, from a columnar database.
4. The method of claim 1 , further comprising:
determining, by the at least one processor, a variation of a third metric of the data over time; and
determining, by the at least one processor, whether a correlation measure between the first metric and the third metric over time, is less than the correlation measure between the first metric and the second metric over time.
5. The method of claim 4 , comprising generating, by the at least one processor responsive to the correlation measure between the first metric and the third metric being less than the correlation measure between the first metric and the second metric, the visualization of the first metric and the second metric varying in time, to display to a user.
6. The method of claim 4 , comprising generating, by the at least one processor responsive to the correlation measure between the first metric and the third metric over time being higher than the correlation measure between the first metric and the second metric over time, a visualization of the first metric and the third metric varying in time, to display to the user.
7. The method of claim 1 , further comprising:
detecting, by the at least one processor, that the data source is a new data source; and
storing, by the at least one processor, the data from the data source over time.
8. The method of claim 1 , further comprising:
detecting, by the at least one processor, that a behavior of the user is indicative of insufficient interest in the visualization of the first metric and the second metric varying in time; and
indicating, by the at least one processor, to refrain from displaying the visualization of the first metric and the second metric at a future time, or to display a different visualization at the future time.
9. The method of claim 8 , wherein the behavior of the user comprises at least one of: dismissing or minimizing the displayed visualization, performing a search unrelated to the first metric or the second metric, or a lack of interaction with the displayed visualization.
10. A device, comprising:
memory configured to store data observed from a data source over time; and
at least one processor configured to:
access data observed from a data source over time;
determine that a variation of at least a first metric or a second metric of the data over time is greater than a threshold;
determine, responsive to determining that the variation is greater than the threshold, that a correlation measure between the first metric and the second metric over time is greater than a predefined reference level; and
select, responsive to the determination that the correlation measure is greater than the predefined reference level, the first metric and the second metric for visualization;
generate, responsive to the selection, a visualization of the first metric and the second metric varying in time, to display to a user;
determine a level of interest for at least one of the first metric or the second metric according to a number of user interactions with the visualization;
modify the correlation measure between the first metric and the second metric based at least on the level of interest; and
update the visualization in accordance with the modified correlation measure.
11. The device of claim 10 , wherein the at least one processor is further configured to:
detect that the data source comprises a new data source; and
dynamically generate, responsive to the new data source, the visualization of the first metric and the second metric varying in time, to display to the user.
12. The device of claim 10 , wherein the memory is configured to store the data in a columnar database; and
wherein the at least one processor is further configured to access the data observed from the data source over time, from the columnar database.
13. The device of claim 10 , wherein the at least one processor is further configured to:
determine a variation of a third metric of the data over time; and
determine whether a correlation measure between the first metric and the third metric over time, is less than the correlation measure between the first metric and the second metric over time.
14. The device of claim 13 , wherein the at least one processor is configured to generate, responsive to the correlation measure between the first metric and the third metric being less than the correlation measure between the first metric and the second metric, the visualization of the first metric and the second metric varying in time, to display to a user.
15. The device of claim 13 , wherein the at least one processor is configured to generate, responsive to the correlation measure between the first metric and the third metric being higher than the correlation measure between the first metric and the second metric, a visualization of the first metric and the third metric varying in time, to display to the user.
16. The device of claim 10 , wherein the reference level comprises a predefined value, or a correlation measure between another pair of metrics of the data.
17. The device of claim 10 , wherein the at least one processor is further configured to:
detect that the data source is a new data source; and
store the data from the data source over time.
18. The device of claim 10 , wherein the at least one processor is further configured to:
detect that a behavior of the user is indicative of insufficient interest in the visualization of the first metric and the second metric varying in time; and
refrain from displaying the visualization of the first metric and the second metric at a future time, or display a different visualization at the future time.
19. A non-transitory computer readable medium storing program instructions for causing one or more processors to:
access data observed from a data source over time;
determine that a variation of at least a first metric or a second metric of the data over time is greater than a threshold;
determine, responsive to determining that the variation is greater than the threshold, that a correlation measure between the first metric and the second metric over time is greater than a reference level; and
select, responsive to determining that the correlation measure is greater than the reference level, the first metric and the second metric for visualization;
generate, responsive to selecting, a visualization of the first metric and the second metric varying in time, to display to a user;
determine a level of interest for at least one of the first metric or the second metric according to a number of user interactions with the visualization;
modify the correlation measure between the first metric and the second metric based at least on the level of interest; and
update the visualization in accordance with the modified correlation measure.