Inferring target clusters based on social connections
View Patent ↗A seed cluster comprising a group of users who share a particular attribute and/or affiliation is determined by a social networking system. For each user of the seed cluster, other users and/or entities connected to the user in the social networking system are retrieved. For each retrieved other user or entity, the social networking system may determine whether the other user or entity exhibits the attribute or affiliation based on a random walk algorithm. A resulting targeting cluster of users and/or entities may be used for targeting advertisements targeting to members. A social networking system may also infer an affiliation for a user based on the user's interaction with a page, application, or entity where other users who interacted with the same page, application, or entity have the same affiliation.
1. A method comprising:
receiving a selection of an attribute shared by a subset of users of an online social networking system;
determining a seed cluster comprising the subset of users sharing the selected attribute;
retrieving a primary connection of a seed user from the seed cluster, the primary connection connected to the seed user in the online social networking system;
determining from stored information about the primary connection whether the retrieved primary connection exhibits the selected attribute;
responsive to a determination that the retrieved primary connection does not exhibit the selected attribute, retrieving a plurality of secondary connections of the seed user in the online social networking system;
determining that one or more of the plurality of secondary connections exhibit the selected attribute;
responsive to a determination that a threshold number of the one or more secondary connections exhibit the selected attribute, inferring, by a processor, that the retrieved primary connection exhibits the selected attribute;
updating the stored information to associate the selected attribute with the retrieved primary connection that was determined to not be exhibited by the primary connection; and
storing information defining a targeting cluster of users as users of the online social networking system who share the selected attribute, where the targeting cluster comprises the seed cluster of users and the primary connection associated with the selected attribute.
2. The method of claim 1 , wherein receiving a selection of an attribute shared by a subset of users of an online social networking system further comprises receiving the selection from an advertiser selecting targeting criteria for an ad request.
3. The method of claim 1 , wherein receiving a selection of an attribute shared by a subset of users of an online social networking system further comprises receiving the selection from a module in the online social networking system.
4. The method of claim 1 , wherein determining a seed cluster comprises adding users to the seed cluster based on an inference related to the selected attribute.
5. The method of claim 4 , wherein adding users to the cluster based on an inference related to the selected attribute comprises adding an inferred user based on retrieved profile information of users connected to the inferred user.
6. The method of claim 1 , wherein the subset of users sharing the selected attribute comprises users of the online social networking system who have recently used a specified application on the online social networking system.
7. The method of claim 1 , wherein determining a seed cluster comprises adding users to the seed cluster based on profile information relating to the selected attribute.
8. The method of claim 1 , wherein determining a seed cluster comprises adding users to the seed cluster based on content information related to the selected attribute.
9. The method of claim 1 , wherein determining that one or more secondary connections exhibit the selected attribute further comprises:
for each secondary connection in a subset of the one or more secondary connections comprising a subset of users of the online social networking system, determining that the secondary connection exhibits the selected attribute based on the user profile of the secondary connection comprising the selected attribute.
10. The method of claim 1 , wherein inferring that the retrieved primary connection exhibits the selected attribute based on the determined one or more secondary connections exhibiting the selected attribute further comprises:
determining a percentage of total secondary connections exhibiting the selected attribute based on the determined one or more secondary connections compared to the retrieved one or more secondary connections connected to the retrieved primary connection in the online social networking system; and
responsive to the determined percentage exceeding a predetermined threshold percentage, inferring that the retrieved primary connection exhibits the selected attribute.
11. The method of claim 10 , wherein the predetermined threshold percentage comprises at least eighty percent.
12. The method of claim 1 , wherein inferring that the retrieved primary connection exhibits the selected attribute based on the determined one or more secondary connections exhibiting the selected attribute further comprises:
determining a weighted average using a plurality of affinity scores of the retrieved primary connection for the determined one or more secondary connections as weights; and
responsive to the determined weighted average exceeding a predetermined threshold, inferring that the retrieved primary connection exhibits the selected attribute.
13. The method of claim 1 , wherein the predetermined criteria comprises at least a predetermined threshold percentage of one or more secondary connections exhibiting the selected attribute.
14. A non-transitory computer readable storage medium storing one or more programs configured to be executed by a server system, the one or more programs comprising instructions for:
receiving a selection of an attribute shared by a subset of users of an online social networking system;
determining a seed cluster comprising the subset of users sharing the selected attribute;
retrieving a primary connection of a seed user from the seed cluster, the primary connection connected to the seed user in the online social networking system;
determining from stored information about the primary connection whether the retrieved primary connection exhibits the selected attribute;
responsive to a determination that the retrieved primary connection does not exhibit the selected attribute, retrieving a plurality of secondary connections of the seed user in the online social networking system;
determining that one or more of the plurality of secondary connections exhibit the selected attribute;
responsive to a determination that a threshold number of the one or more secondary connections exhibit the selected attribute, inferring, by a processor, that the retrieved primary connection exhibits the selected attribute;
updating the stored information to associate the selected attribute with the retrieved primary connection that was determined to not be exhibited by the primary connection; and
storing information defining a targeting cluster of users as users of the online social networking system who share the selected attribute, where the targeting cluster comprises the seed cluster of users and the primary connection associated with the selected attribute.
15. The non-transitory computer readable storage medium of claim 14 , wherein receiving a selection of an attribute shared by a subset of users of an online social networking system further comprises receiving the selection from an advertiser selecting targeting criteria for an ad request.
16. The non-transitory computer readable storage medium of claim 14 , wherein receiving a selection of an attribute shared by a subset of users of an online social networking system further comprises receiving the selection from a module in the online social networking system.
17. The non-transitory computer readable storage medium of claim 14 , wherein determining a seed cluster comprises adding users to the seed cluster based on an inference related to the selected attribute.
18. The non-transitory computer readable storage medium of claim 17 , wherein adding users to the cluster based on an inference related to the selected attribute comprises adding an inferred user based on retrieved profile information of users connected to the inferred user.
19. The non-transitory computer readable storage medium of claim 14 , wherein the subset of users sharing the selected attribute comprises users of the online social networking system who have recently used a specified application on the online social networking system.
20. The non-transitory computer readable storage medium of claim 14 , wherein determining a seed cluster comprises adding users to the seed cluster based on profile information relating to the selected attribute.
21. The non-transitory computer readable storage medium of claim 14 , wherein determining a seed cluster comprises adding users to the seed cluster based on content information related to the selected attribute.
22. The non-transitory computer readable storage medium of claim 14 , wherein determining that one or more secondary connections exhibit the selected attribute further comprises:
for each secondary connection in a subset of the one or more secondary connections comprising a subset of users of the online social networking system, determining that the secondary connection exhibits the selected attribute based on the user profile of the secondary connection comprising the selected attribute.
23. The non-transitory computer readable storage medium of claim 14 , wherein inferring that the retrieved primary connection exhibits the selected attribute based on the determined one or more secondary connections exhibiting the selected attribute further comprises:
determining a percentage of total secondary connections exhibiting the selected attribute based on the determined one or more secondary connections compared to the retrieved one or more secondary connections connected to the retrieved primary connection in the online social networking system; and
responsive to the determined percentage exceeding a predetermined threshold percentage, inferring that the retrieved primary connection exhibits the selected attribute.
24. The non-transitory computer readable storage medium of claim 23 , wherein the predetermined threshold percentage comprises at least eighty percent.
25. The non-transitory computer readable storage medium of claim 14 , wherein inferring that the retrieved primary connection exhibits the selected attribute based on the determined one or more secondary connections exhibiting the selected attribute further comprises:
determining a weighted average using a plurality of affinity scores of the retrieved primary connection for the determined one or more secondary connections as weights; and
responsive to the determined weighted average exceeding a predetermined threshold, inferring that the retrieved primary connection exhibits the selected attribute.
26. The non-transitory computer readable storage medium of claim 14 , wherein the predetermined criteria comprises at least a predetermined threshold percentage of one or more secondary connections exhibiting the selected attribute.