Sponsored recommendation in a social networking system

Patent metadata

Field	Value
Publication number	US-10475134-B2
Application number	US-201313749557-A
Country	US
Kind code	B2
Filing date	Jan 24, 2013
Priority date	Jan 24, 2013
Publication date	Nov 12, 2019
Grant date	Nov 12, 2019

Abstract

A social networking system presents suggestions to a user of a social networking system to use or install one or more applications accessible in the system. The social networking system offers suggestions by ranking candidate applications for a particular user and selecting applications to recommend based on the rankings. Advertisers associated with some applications may bid to boost the rankings of an application, making it more likely to appear for a given user.

First claim

What is claimed is:

1. A method comprising:
identifying a user of a social networking system;
maintaining, at the social networking system, a plurality of data stores including a content data store comprising a plurality of objects and an edge data store comprising plurality of connections therebetween;
receiving, at the social networking system, a plurality of recommendation units for the user, each recommendation unit including a link to a recommendation that suggests the user perform an action in the social networking system on an object in the content data store identified by the recommendation unit, the action performed by the user, responsive to the user accessing the link, being stored in the edge data store by the social networking system;
retrieving, from the edge data store, information describing one or more connections in the edge data store between the user and the object identified by each of the recommendation units;
determining the user's prior interactions with the object identified by each recommendation unit based on the retrieved information describing one or more connections in the edge data store between the user and the object identified by each of the recommendation units;
computing a score in a first dimension for each of the plurality of recommendation units based at least in part on the determined prior interactions of the user with the object identified by each recommendation unit;
receiving a score in a second dimension from a content provider associated with a sponsored recommendation unit of the plurality of recommendation units;
converting at least the score in the first dimension or the score in the second dimension into scores in a common dimension for comparing the scores in the first dimension and the scores in the second dimension;
increasing a score in the common dimension for the sponsored recommendation unit relative to one or more of the other recommendation units based on the received score from the content provider associated with the sponsored recommendation unit;
ranking the plurality of recommendation units based on the scores in the common dimension;
selecting one or more recommendation units based at least in part on the ranking;
organizing a user interface to include the selected one or more recommendation units, wherein the selected one or more recommendation units are ordered within a limited display space in the user interface according to the ranking; and
sending the organized user interface for display in a computing device of the user.

2. The method of claim 1 , wherein the score of one or more of the recommendation units is based at least in part on an amount of interaction between the user and the social networking system.

3. The method of claim 1 , wherein ranking the plurality of recommendation units comprises:
determining, in association with a sponsored recommendation unit, an expected-value score in the second dimension, wherein the expected-value score is based in part on the received score in the second dimension from the content provider;
converting the expected-value score into the common dimension;
modifying a score in the common dimension associated with the sponsored recommendation unit based on the expected-value score in the common dimension; and
ranking the recommendation units based at least in part on the scores.

4. The method of claim 3 , wherein the expected-value score is based on the received score in the second dimension from the content provider and a likelihood of the user interacting with the sponsored recommendation unit.

5. The method of claim 4 , wherein modifying the score in the common dimension associated with the sponsored recommendation unit based on the expected-value score in the common dimension comprises:
increasing the score in the common dimension associated with the sponsored recommendation unit by an amount determined in part by the expected-value score.

6. A method comprising:
identifying a user of a social networking system;
maintaining, at the social networking system, a plurality of data stores including a content data store comprising a plurality of objects and an edge data store comprising plurality of connections therebetween;
receiving, at the social networking system, a plurality of recommendation units for the user, each recommendation unit including a link to a recommendation that suggests the user perform an action in the social networking system on an object in the content data store identified by the recommendation unit, the action performed by the user, responsive to the user accessing the link, being stored in the edge data store by the social networking system;
retrieving, from the edge data store, information describing one or more connections in the edge data store between the user and the object identified by each of the recommendation units;
determining, for an additional user of the social networking system, the additional user's prior interactions with the object identified by each recommendation unit based on the retrieved information describing one or more connections in the edge data store between the user and the object identified by each recommendation unit;
computing a score in a first dimension for each of the plurality of recommendation units based at least in part on the determined prior interactions of the additional users of the social networking system with the object identified by each recommendation unit;
receiving a score in a second dimension from a content provider associated with one or more sponsored recommendation units of the plurality of recommendation units;
converting at least the score in the first dimension or the score in the second dimension into scores in a common dimension;
increasing a score in the common dimension for each sponsored recommendation unit based on the received score from the content provider associated with the sponsored recommendation unit;
ranking the plurality of recommendation units based on the scores in the common dimension;
selecting one or more recommendation units based at least in part on the ranking;
organizing a user interface to include the selected one or more recommendation units, wherein the selected one or more recommendation units are ordered within a limited display space in the user interface according to the ranking; and
sending the organized user interface for display in a computing device of the user.

7. The method of claim 6 , wherein the score of a recommendation unit is based at least in part on an amount of interaction between the user and the social networking system.

8. The method of claim 6 , wherein ranking the plurality of recommendation units comprises:
determining, in association with a sponsored recommendation unit, an expected-value score in the second dimension, wherein the expected-value score is based at least in part on the received score in the second dimension from the content provider;
converting the expected-value score into the common dimension;
modifying a score in the common dimension associated with each sponsored recommendation unit based on the expected-value score in the common dimension; and
ranking the recommendation units based at least in part on the scores.

9. The method of claim 8 , wherein the expected-value score is based on the received score in the second dimension from the content provider and a likelihood of the user interacting with the sponsored recommendation unit.

10. The method of claim 8 , wherein modifying the score in the common dimension associated with the sponsored recommendation unit based on the expected-value score in the common dimension comprises:
increasing the score in the common dimension associated with a sponsored recommendation

Assignees

• Facebook Inc

Inventors

• Liu Deborah
• Mathur Nipun
• Kanter Jeffrey Andrew
• Raji Vijaye

Classifications

• G06F16/951

  Indexing; Web crawling techniques · CPC title

• G06Q30/0277 Primary

  Online advertisement · CPC title

• G06Q10/40

  Business processes related to social networking or social networking services · CPC title

• G06Q50/01 Primary

  Physics · mapped topic

• G06Q10/48

  using social graphs · CPC title