Determining object relevance in a social networking system

• US 9,325,751 B2
• Filed: 11/28/2012
• Issued: 04/26/2016
• Est. Priority Date: 11/28/2012
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Bosworth, Andrew G., Bharathi, Aruna
• Primary Examiner(s): Donabed, Ninos
• Application Number:
  US13/688,020
• Publication Number:
  US 20140149504A1
• Time in Patent Office: 1,245 Days
• Field of Search: None
• US Class Current: 1/1
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 65/403 Arrangements for multi-party communication, e.g. for conferences data switching systems for conference H04L12/18; arrangements for connecting several subscribers to a common circuit, i.e. affording conference facilities H04M3/56; television conferencing systems H04N7/15

Abstract

An influence metric describing the influence of a social networking system object on social networking system users is determined based on affinities between the users and the object. For example, affinities between the associated users and the object are combined to determine the influence metric. Content may be selected for presentation to users based in part on influence metrics of the content. Additionally, influence metrics of objects associated with a user may be combined to determine the relevance of objects associated with the user, which may also be used to select content for presentation to the user.

Claims

1. A method comprising:

storing associations between a plurality of objects and a plurality of users in a social networking system;

identifying a target user of the social networking system;

identifying objects associated with the target user based on the stored associations;

retrieving an influence metric associated with each of the identified objects, each influence metric identifying the influence of an associated object on a plurality of users associated with the object based on actions performed by the plurality of users related to the associated object and based on an amount of time since each action was performed such that an influence metric is weighted more heavily by a first action than by a second action, the first action performed more recently than the second action;

generating, by a hardware processor, a relevance metric for the target user based on an aggregation of the retrieved influence metrics, the relevance metric describing an aggregated relevance of the identified objects with which the target user is associated in the social networking system;

selecting content for presentation based on an influence metric associated with the selected content;

selecting the target user from the plurality of users based on the relevance metric for the target user; and

presenting the selected content to the target user as a story in a social networking system newsfeed interface at a location within the newsfeed interface based on the influence metric associated with the selected content, the newsfeed interface displaying stories ordered based on influence metrics associated with objects associated with the stories.

2. The method of claim 1, wherein generating the relevance metric for the target user based on the aggregation of the retrieved influence metrics comprises:

determining a median or a mean of the influence metrics.

3. The method of claim 1, wherein generating the relevance metric for the target user based on the aggregation of the retrieved influence metrics comprises:

applying a weight to each of the influence metrics to generate weighted influence metrics; and

generating the relevance metric based on the weighted influence metrics.

4. The method of claim 3, wherein a weight applied to an influence metric is based at least in part on a characteristic of an object associated with an influence metric.

5. The method of claim 4, wherein the characteristic is a type associated with the object or an affinity between the target user and the object.

6. The method of claim 3, wherein generating the relevance metric based on the weighted influence metrics comprises:

determining a median or a mean of the weighted influence metrics.

7. The method of claim 1, wherein identifying objects associated with the target user based on the stored associations comprises:

identifying objects associated with the target user and having a specified characteristic.

8. The method of claim 7, wherein the specified characteristic is selected from a group consisting of:

a type of object, an affinity between the target user and an object, and any combination thereof.

9. The method of claim 1, wherein selecting content comprises selecting a content object of a first type for presentation to the target user if the relevance metric for the target user is less than a threshold, and selecting a content object of a second type for presentation to the target user if the relevance metric for the target user is greater than the threshold.

10. A method comprising:

identifying a target user of a social networking system;

identifying objects associated with the target user based on associations between a plurality of objects and the target user stored by the social networking system;

retrieving an influence metric associated with each of the identified objects, each influence metric identifying an influence of an associated object on a plurality of users associated with the object based on actions performed by the plurality of users related to the associated object and based on an amount of time since each action was performed such that an influence metric is weighted more heavily by a first action than by a second action, the first action performed more recently than the second action;

generating, by a hardware processor, a relevance metric for the target user based on the retrieved influence metrics, the relevance metric describing an aggregated relevance of the identified objects in the social networking system; and

selecting content for presentation to the target user based on an influence metric associated with the selected content;

selecting the target user from the plurality of users based on the relevance metric for the target user; and

presenting the selected content to the target user as a story in a social networking system newsfeed interface at a location within the newsfeed interface based on the influence metric associated with the selected content, the newsfeed interface displaying stories ordered based on influence metrics associated with objects associated with the stories.

11. The method of claim 10, wherein generating the relevance metric for the target user based on the retrieved influence metrics comprises:

determining a median or a mean of the influence metrics.

12. The method of claim 10, wherein generating the relevance metric for the target user based on the retrieved influence metrics comprises:

applying a weight to each of the influence metrics to generate weighted influence metrics; and

generating the relevance metric based on the weighted influence metrics.

13. The method of claim 10, wherein the content includes at least one advertisement.

14. The method of claim 10, wherein identifying objects associated with the target user based on the stored associations comprises:

identifying objects associated with the target user and having a specified characteristic.

15. The method of claim 10, wherein the specified characteristic is selected from a group consisting of:

a type of object, an affinity between the target user and an object, and any combination thereof.

16. A method comprising:

identifying a plurality of users of a social networking system, each of the plurality of users having a common characteristic;

identifying objects associated with each of user of the plurality of users based on associations between the plurality of users and the identified objects maintained by the social networking system;

retrieving an influence metric associated with each of the identified objects, each influence metric for an associated object comprising a measure of an influence of the object on a plurality of users associated with the object based on actions performed by the plurality of users related to the associated object and based on an amount of time since each action was performed such that an influence metric is weighted more heavily by a first action than by a second action, the first action performed more recently than the second action;

generating, by a hardware processor, a relevance metric for each user of the plurality of users, the relevance metric for a user based on the influence metrics of the identified objects associated with the user; and

generating an aggregated relevance metric for the plurality of users based on the relevance metrics of the plurality of users;

selecting content for presentation based on an influence metric associated with the selected content and based on the aggregated relevance metric; and

presenting the selected content to the plurality of users as a story in a social networking system newsfeed interface at a location within the newsfeed interface based on the influence metric associated with the selected content, the newsfeed interface displaying stories ordered based on influence metrics associated with objects associated with the stories.

17. The method of claim 16, wherein generating the aggregated relevance metric for the plurality of users based on the relevance metrics of the plurality of users comprises:

determining a median or a mean of the relevance metrics.

18. The method of claim 16, wherein generating the aggregated relevance metric for the plurality of users based on the relevance metrics of the plurality of users comprises:

applying a weight to each of the relevance metrics to generate weighted relevance metrics; and

generating the aggregate relevance metric based on the weighted relevance metrics.

19. The method of claim 16, wherein the common characteristic is a geographical region associated with the identified plurality of users.

20. The method of claim 16, further comprising:

selecting content for presentation to one or more of the plurality of users based at least in part on the aggregated relevance metric for the plurality of users.