App store peer group benchmarking with differential privacy
Providing peer group benchmarking with differential privacy obtaining one or more app store metrics for a first application, determining an application peer group for the first application, wherein the application peer group is determined based on a plurality of common traits, and obtaining one or more peer group app store metrics for the application peer group based on the one or more app store metrics. A user interface is displayed to indicate a relative placement of at least one of the one or more app store metrics for the first application among the peer group app store metrics, and the user interface element identifies the application peer group metrics with a minimum level of accuracy without revealing the performance of individual apps within the peer group.
1. A method comprising:
determining, for a first application, an application peer group based on a plurality of common traits among a set of app store applications, wherein the plurality of common traits are selected to cause a size of the application peer group to satisfy a privacy threshold;
privatizing the application peer group to cause the application peer group to satisfy the privacy threshold, wherein the privatizing injects a margin of error into one or more peer group app store metrics;
confirming that the privatized application peer group satisfies an accuracy threshold;
obtaining the one or more peer group app store metrics for the privatized application peer group and one or more app store metrics for the first application; and
causing a user interface element to be displayed to indicate a relative value of at least one of the one or more app store metrics for the first application among the one or more peer group app store metrics,
wherein the user interface element identifies the application peer group without revealing a performance of individual apps within the application peer group.
2. The method of claim 1 , wherein determining the application peer group comprises:
determining a category set for the first application; and
performing a match using the category set for the first application and a category set for each member of the application peer group.
3. The method of claim 2 , wherein the category set for the first application comprises at least one selected from a group consisting of an application type, a monetization type, and a download volume.
4. The method of claim 1 , further comprising:
in accordance with a determination that the application peer group fails to satisfy the accuracy threshold:
discarding at least one of the plurality of common traits to obtain a set of remaining common traits, and
revising the application peer group based on the set of remaining common traits.
5. The method of claim 1 , wherein privatizing the application peer group obfuscates individual peer group app store metrics among the application peer group.
6. The method of claim 1 , wherein the one or more peer group app store metrics comprise at least one selected from a group consisting of a discoverability metric, a usage metric, and a monetization metric.
7. The method of claim 1 , wherein the user interface element identifies the application peer group without revealing members of the application peer group.
8. A non-transitory computer readable medium comprising computer readable code executable by one or more processors to:
determine, for a first application, an application peer group based on a plurality of common traits among a set of app store applications, wherein the plurality of common traits are selected to cause a size of the application peer group to satisfy a privacy threshold;
privatizing the application peer group to cause the application peer group to satisfy the privacy threshold, wherein the privatizing injects a margin of error into one or more peer group app store metrics;
confirming that the privatized application peer group satisfies an accuracy threshold;
obtain the one or more peer group app store metrics for the application peer group and one or more app store metrics for the first application; and
cause a user interface element to be displayed to indicate a relative value of at least one of the one or more app store metrics for the first application among the one or more peer group app store metrics,
wherein the user interface element identifies the application peer group without revealing a performance of individual apps within the application peer group.
9. The non-transitory computer readable medium of claim 8 , wherein the computer readable code to determine the application peer group comprises computer readable code to:
determine a category set for the first application; and
perform a match using the category set for the first application and a category set for each member of the application peer group.
10. The non-transitory computer readable medium of claim 9 , wherein the category set for the first application comprises at least one selected from a group consisting of an application type, a monetization type, and a download volume.
11. The non-transitory computer readable medium of claim 8 , further comprising computer readable code to:
determine that the application peer group fails to satisfy a predetermined accuracy threshold and,
in accordance with the determination that the application peer group fails to satisfy the accuracy threshold:
discard at least one of the plurality of common traits to obtain a set of remaining common traits, and
revise the application peer group based on the set of remaining common traits.
12. The non-transitory computer readable medium of claim 8 , wherein privatizing the application peer group obfuscates individual peer group app store metrics among the application peer group.
13. The non-transitory computer readable medium of claim 8 , wherein the one or more peer group app store metrics comprise at least one selected from a group consisting of a discoverability metric, a usage metric, and a monetization metric.
14. The non-transitory computer readable medium of claim 8 , wherein the user interface element identifies the application peer group without revealing members of the application peer group.
15. A system comprising:
one or more processors; and
one or more non-transitory computer readable media comprising computer readable code executable by the one or more processors to:
determine, for a first application, an application peer group based on a plurality of common traits among a set of app store applications, wherein the plurality of common traits are selected to cause a size of the application peer group to satisfy a privacy threshold;
privatizing the application peer group to cause the application peer group to satisfy the privacy threshold, wherein the privatizing injects a margin of error into one or more peer group app store metrics;
confirming that the privatized application peer group satisfies an accuracy threshold; and
obtain the one or more peer group app store metrics for the application peer group and one or more app store metrics for the first application; and
cause a user interface element to be displayed to indicate a relative value of at least one of the one or more app store metrics for the first application among the one or more peer group app store metrics,
wherein the user interface element identifies the application peer group without revealing a performance of individual apps within the application peer group.
16. The system of claim 15 , wherein the computer readable code to determine the application peer group comprises computer readable code to:
determine a category set for the first application; and
perform a match using the category set for the first application and a category set for each member of the application peer group.
17. The system of claim 15 , further comprising computer readable code to:
determine that the application peer group fails to satisfy a predetermined accuracy threshold and,
in accordance with the determination that the application peer group fails to satisfy the accuracy threshold:
discard at least one of the plurality of common traits to obtain a set of remaining common traits, and
revise the application peer group based on the set of remaining common traits.
18. The system of claim 15 , wherein privatizing the application peer group obfuscates individual peer group app store metrics among the application peer group.
19. The system of claim 15 , wherein the one or more peer group app store metrics comprise at least one selected from a group consisting of a discoverability metric, a usage metric, and a monetization metric.
20. The system of claim 15 , wherein the user interface element identifies the application peer group without revealing members of the application peer group.