Suggesting Connections to a User Based on an Expected Value of the Suggestion to the Social Networking System

• US 20120041907A1
• Filed: 08/16/2010
• Published: 02/16/2012
• Est. Priority Date: 08/16/2010
• Status: Active Grant

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Wang, James, Ferrante, Daniel, Burge, Jen, Backstrom, Lars Seren, Ratiu, Florin
• Granted Patent: US 8,521,661 B2
• Time in Patent Office: Days
• US Class Current: 706/12
• CPC Class Codes: G06F 15/163 Interprocessor communication
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/3322 using system suggestions G06F16/3325 takes precedence
• CPC Class Codes: G06N 5/02 Knowledge representation; Symbolic representation
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising

Abstract

To suggest new connections to a user of a social networking system, the system generates a set of candidate users to whom the user has not already formed a connection. The system determines the likelihood that the user will connect to each candidate user if suggested to do so, and it also computes the value to the social networking system if the user does connect to the candidate user. Then, the system computes an expected value score for each candidate user based on the corresponding likelihood and the value. The candidate users are ranked and the suggestions are provided to the user based on the candidate users'"'"' expected value scores. The social networking system can suggest other actions to a user in addition to forming a new connection with other users.

Claims

15. A method for suggesting an action to a user of a social networking system, the method comprising:

generating a list of actions comprising candidate actions available to a user of the social networking system;

predicting the likelihood that a user will perform each candidate action, the likelihood calculated responsive to historical data associated with the user within the social networking system;

determining a value to the social networking system of the user performing each candidate action;

computing an expected value for each candidate action, the expected value based on the predicted likelihood of the user performing the candidate action and the determined value to the social networking system of the user performing the candidate action; and

suggesting one or more candidate actions to the user, the candidate action suggested based on the computed expected value associated with each candidate action.

1. A method for suggesting new connections to a user of a social networking system, the method comprising:

generating a list of candidate connections not connected to the user in the social networking system;

predicting the likelihood of a connection between the user and each candidate connection, the likelihood calculated responsive to historical user data within the social networking system;

determining a value to the social networking system of a new connection between the user and each candidate connection, the value based on data associated with the user and the connection in the social networking system;

computing an expected value for each candidate connection, the expected value based on the predicted likelihood of a connection between the user and the candidate connection and the determined value to the social networking system of a new connection between the user and the candidate connection; and

suggesting one or more candidate connections to the user, the candidate connections suggested based on the computed expected value associated with each candidate connection.

2. The method of claim 1, wherein candidate connections include objects, entities, concepts or other users not connected with the user within the social networking system.

3. The method of claim 1, wherein predicting the likelihood comprises using a machine learning algorithm, the machine learning algorithm trained using historical data collected about user interactions in the social networking system.

4. The method of claim 1, wherein the predicting the likelihood is based on one or more of the number of mutual connections between the user and the candidate connection, including the number of mutual connections, the ratio of mutual connections to total connections, time discounted mutual friends, common networks, gender, high school, college, graduation year, interests, favorite books, movies, television shows, music, being identified in the same pictures, the number of times the user was displayed a picture of each candidate connection, the number of times the user clicked on a link of each candidate connection.

5. The method of claim 1, wherein determining the value to the social networking system of a new connection between the user and each candidate connection is based at least in part on a number of existing connections of the candidate connection, wherein the value of a new connection is higher when the number of existing connections is lower.

6. The method of claim 1, wherein determining the value to the social networking system of a new connection between the user and each candidate connection is based at least in part on a frequency of social networking system usage of the candidate connection, wherein the value of a new connection is higher when the frequency of social networking system usage is lower.

7. The method of claim 1, wherein determining the value to the social networking system of a new connection between the user and each candidate user is based at least in part on a time since joining the social networking system of the candidate user, wherein the value of a new connection is higher when the time since joining the social networking system is lower.

8. The method of claim 1, wherein determining the value to the social networking system of a new connection between the user and each candidate user is based at least in part on demographic information about the candidate user.

9. The method of claim 1, wherein the one or more suggestions are provided to the user in a page of information, without the user requesting the suggestions.

10. The method of claim 1, wherein the one or more suggestions are provided to the user responsive to a query from the user for suggestions.

11. The method of claim 1, further comprising discounting the expected values associated with each of the candidate users suggested to the user.

12. The method of claim 1, further comprising:

discounting the expected values associated with each of the candidate connection suggested to the user; and

suggesting one or more candidate connection to the user, the suggestions based on the discounted expected values associated with each of the candidate users.

13. The method of claim 11, wherein discounting the expected value associated with each candidate connection is a function of the number of times the candidate connection has been previously suggested to the user.

14. The method of claim 1, further comprising:

selecting one or more additional candidate connections to suggest to the user to add as a new connection in the social networking system based on discounted expected values; and

providing the additional suggestions to the user.

16. The method of claim 15, wherein the list of actions comprises candidate actions the user can perform within the social networking system and a list of connections, the connections capable of interacting with the user through the candidate actions.

18. The method of claim 15, wherein the candidate action comprises of playing a video game within the social networking system and the candidate list of actions comprise a list of video games available for the user to play within the social networking system.

19. The method of claim 15, wherein predicting the likelihood comprises using a machine learning algorithm, the machine learning algorithm analyzing historical, behavioral and interaction data of users of the social networking system to predict the probability of a user performing a particular action within the social networking system.

20. The method of claim 15, wherein determining the value to the social networking system of a new action suggested to the user is based at least in part on a number of prior actions performed by the user, wherein the value of a suggested action is higher when the number of existing actions is lower.

21. The method of claim 15, wherein determining the value to the social networking system of a new action suggested to the user is based at least in part on a number of existing connections of the user or the connection, wherein the value of a suggested action is higher when the number of existing connections is lower.

22. The method of claim 15, wherein determining the value to the social networking system of a new action suggested to the user is based at least in part on a frequency of social networking system usage of the user, wherein the value of a suggested action is higher when the frequency of social networking system usage is lower.

23. The method of claim 15, wherein determining the value to the social networking system of a new action suggested to the user is based at least in part on a time since joining the social networking system of the user, wherein the value of a suggested action is higher when the time since joining the social networking system is lower.

24. The method of claim 15, wherein determining the value to the social networking system of a new action suggested to the user is based at least in part on demographic information about the user.

25. The method of claim 15, wherein the one or more suggestions are provided to the user in a page of information, without the user requesting the suggestions.

26. The method of claim 15, wherein the one or more suggestions are provided to the user responsive to a query from the user for suggestions.

27. The method of claim 15, further comprising discounting the expected values associated with each of the candidate actions suggested to the user.

28. The method of claim 15, further comprising

discounting the expected values associated with each of the candidate actions suggested to the user;

and

suggesting one or more candidate actions to the user, the suggestions based on the discounted expected values associated with each of the candidate actions.

29. The method of claim 15, wherein discounting the expected value associated with each candidate action is a function of the number of times the candidate action has been previously suggested to the user.

16-1. The method of claim 15, wherein the candidate action comprises an action selected from a group comprising sending a message to a connection, sending an instant message to a connection, posting text on a connection'"'"'s profile page, sending a virtual gift to a connection, commenting on a connection'"'"'s activity or broadcast post, commenting on a connection'"'"'s pictures, playing a particular game among a group of games, updating the user'"'"'s profile page, replying to an invitation, becoming a fan of a page, and inviting another user to a group.