System and method for invitation targeting in a web-based social network

• US 20090049127A1
• Filed: 08/16/2007
• Published: 02/19/2009
• Est. Priority Date: 08/16/2007
• Status: Abandoned Application

• Current Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Original Assignee: Meta Platforms, Inc. (f/k/a Facebook, Inc.)
• Inventors: Juan, Yun-Fang, Jin, Kang-Xing
• Application Number:
  US11/893,797
• Publication Number:
  US 20090049127A1
• Time in Patent Office: Days
• US Class Current: 709/204
• CPC Class Codes: G06Q 10/10 Office automation; Time management
• CPC Class Codes: H04L 67/306 User profiles

Abstract

A system and method for selecting users of a web-based social network who are likely to respond to an invitation, each of the users having associated profile information is disclosed. The method includes selecting pilot users and a reduced set of keywords from the profile information. The method further includes sending the invitation to the pilot users, receiving responses from the pilot users, and classifying the responses as either positive or negative. A training set of vector pairs is created each vector pair representing a pilot user and including data representing a classified response and training keywords selected from the reduced set of keywords and associated profile information for the pilot user. A function is determined based on the vectors and used to calculate a likelihood that each of one or more users of the web based social network will respond to the invitation.

Claims

1. A method for selecting users of a web-based social network who are likely to respond to an invitation, each of the users having associated profile information, the method comprising:

selecting a plurality of pilot users from the users in the web based social network;

selecting a reduced set of keywords from the profile information for the pilot users;

sending the invitation to the pilot users;

receiving responses to the invitation from the pilot users;

classifying the received responses as either positive or negative;

creating a training set of vector pairs, each vector pair representing a pilot user and including data representing the classified response received from the pilot user and a set of training keywords selected from the reduced set of keywords and based at least in part on the associated profile information for the pilot user;

determining a function based on the training set of vector pairs that calculates from a user'"'"'s profile information a likelihood that the user will respond to the invitation; and

calculating from the function a likelihood that each of one or more of the users in the web based social network will respond to the invitation.

2. The method of claim 1, further comprising ranking the one or more users based on the calculated likelihood of each user responding and sending the invitation to a predetermined number of users having the highest calculated likelihood of responding.

3. The method of claim 1, further comprising sending the invitation to users having a calculated likelihood of responding that is higher than a threshold value.

4. The method of claim 1, wherein selecting a plurality of pilot users comprises randomly selecting a predetermined number of pilot users from the users of the web based social network.

5. The method of claim 1, wherein selecting a reduced set of keywords comprises using singular vector decomposition (SVD).

6. The method of claim 1, wherein selecting a reduced set of keywords comprises using latent dirichlet allocation.

7. The method of claim 1, wherein selecting a reduced set of keywords comprises using probabilistic latent semantic indexing (PLSI).

8. The method of claim 1, wherein selecting a reduced set of keywords comprises using feature selection.

9. The method of claim 1, wherein determining a function based on the training set of vector pairs comprises using a boosted trees method.

10. The method of claim 1, wherein determining a function based on the training set of vector pairs comprises using support vector machine (SVM) analysis.

11. The method of claim 1, wherein determining a function based on the training set of vector pairs comprises using neural network analysis.

12. The method of claim 1, wherein classifying the received responses comprises classifying the received responses as positive if the user clicks on the invitation.

13. The method of claim 1, wherein classifying the received responses comprises classifying the received responses as positive if the user requests information about the invitation.

14. The method of claim 1, wherein classifying the received responses comprises classifying the received responses as positive if the user makes a purchase.

15. The method of claim 1, wherein classifying the received responses comprises classifying the received response as negative if the user does not respond to the invitation within a certain time.

16. The method of claim 1, wherein creating a training set of vectors for one or more pilot users further comprises:

stratifying the pilot users based on classified responses; and

randomly sampling the stratified pilot users.

17. The method of claim 1, wherein creating a training set of vector pairs further comprises:

selecting a first set of pilot users;

selecting a second set of pilot users;

determining a scalar based on a number of pilot users in the first set and a number of pilot users in the second set; and

multiplying each vector pair representing pilot users in the first set by the scalar.

18. The method of claim 17, wherein the scalar is about equal to a number of pilot users in the second set divided by a number of pilot users in the first set.

19. The method of claim 17, wherein the first set of pilot users is selected from pilot users providing a positive response and the second set of pilot users is selected from pilot users providing a negative response.

20. The method of claim 17, wherein the first set of pilot users is selected from pilot users providing a positive response and the second set of pilot users is selected from a random sample of all pilot users.

21. The method of claim 17, wherein the first set of pilot users is selected from pilot users providing a positive response and the second set of pilot users is selected from a random sample of pilot users providing a negative response.

22. The method of claim 1, wherein sending the invitation to a pilot user comprises embedding the invitation into a personal web page.