Detecting content on a social network using browsing patterns

• US 9,338,047 B1
• Filed: 10/05/2012
• Issued: 05/10/2016
• Est. Priority Date: 10/01/2009
• Status: Active Grant

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: Google Inc. (Alphabet Inc.)
• Inventors: Kunal, Ranveer, Panesar, Kiran S., Thakur, Madhukar N., Asgekar, Amogh S.
• Primary Examiner(s): Ouellette, Jonathan
• Application Number:
  US13/646,159
• Time in Patent Office: 1,313 Days
• Field of Search: 705 11-912
• US Class Current: 1/1
• CPC Class Codes: G06Q 30/0201 Market modelling; Market analysis; Collecting market data
• CPC Class Codes: G06Q 50/01 Social networking
• CPC Class Codes: H04L 67/535 Tracking the activity of the user network monitoring arrangements H04L43/00; recording of computer activity G06F11/34

Abstract

In one implementation, a computer-implemented method includes receiving information related to users of a social network site and determining a graph for at least a portion of the social network, the graph including a plurality of nodes connected by directional links where each node corresponds to a user having a profile page on the social network and where each directional link corresponds to a user viewing at least an aspect of a profile page for another user. The method can further include identifying first nodes as including content associated with a particular subject of interest, and seeding the identified first nodes with first scores. The method can also include determining second scores for second nodes by propagating the first scores across the social graph based upon third nodes having viewed aspects of profile pages associated with the first nodes and the second nodes; and providing the determined second scores.

Claims

1. A computer-implemented method comprising:

obtaining, by a computer system, information that identifies viewing activity of users with regard to content that is associated with social network profiles, wherein the viewing activity identifies content that the users have viewed;

identifying, by the computer system, a portion of the social network profiles, associated with a first group of users, that include a particular category of content;

assigning, to the portion of the social network profiles, initial scores that indicate that the portion of the social network profiles include the particular type of content;

determining, by the computer system, viewer scores for the first group of users based on i) the viewing activity of the first group of users and ii) the initial scores that are assigned to the portion of the social network profiles;

determining, by the computer system, content scores for each user in a second group of users in a remaining portion of the social network profiles based on, for each user in the second group, i) the viewing activity of a particular user in the second group of users, ii) the viewer scores for the first group of users, and iii) a weighting value that indicates a closeness of a relationship between the particular user in the second group of users and a particular user in the first group of users; and

selecting, by the computer system and based on the content scores, one or more users in the second group of users in the remaining portion of the social network profiles as having at least a threshold likelihood of including the particular category of content.

2. The computer-implemented method of claim 1, wherein a particular viewer score for a particular user is determined based on a combination of scores for particular social network profiles that the particular user has viewed, as indicated by the viewing activity.

3. The computer-implemented method of claim 2, wherein the combination of scores for the particular social network profiles comprises a mean of the scores for the particular social network profiles.

4. The computer-implemented method of claim 2, wherein the combination of scores for the particular social network profiles comprises a root mean square of the scores for the particular social network profiles.

5. The computer-implemented method of claim 2, wherein a particular content score for a particular social network profile is determined based on a combination of particular viewer scores for particular users who have viewed the particular social network profile, as indicated by the viewing activity.

6. The computer-implemented method of claim 5, wherein the combination of particular viewer scores for the particular users comprises a mean of the particular viewer scores for the particular users.

7. The computer-implemented method of claim 5, further comprising:

determining the weighting value for the particular users based on a number of the particular users with viewing scores that are greater than or equal to a threshold score; and

wherein the combination of particular viewer scores for the particular users comprises a mean of the particular viewer scores for the particular users that is weighted by the weighting value for the particular users.

8. The computer-implemented method of claim 7, wherein the weighting value comprises a log of the number of particular users plus one.

9. The computer-implemented method of claim 5, wherein the combination of the particular viewer scores for the particular users comprises a root mean square of the particular viewer scores for the particular users.

10. The computer-implemented method of claim 1, wherein the particular category of content comprises at least one of the group consisting of adult content and illegal content.

11. The computer-implemented method of claim 1, further comprising:

iteratively determining the viewer scores and determining the content scores for at least a threshold number of iterations; and

wherein the one or more of the remaining portion of the social network profiles is selected based on the content scores after at least the threshold number of iterations.

12. A computer system comprising:

one or more computing devices;

an interface of the one or more computing devices to obtain information that identifies viewing activity of users with regard to content that is associated with social network profiles, wherein the viewing activity identifies content that the users have viewed;

a score seeding component that is programmed to identify a portion of the social network profiles, associated with a first group of users, that include a particular category of content, and to assign, to the portion of the social network profiles, initial scores that indicate that the portion of the social network profiles include the particular type of content;

a score propagation module that is programmed to determine viewer scores for the first group of users based on i) the viewing activity of the users and ii) the initial scores that are assigned to the portion of the social network profiles, and to determine content scores for each user in a second group of users in a remaining portion of the social network profiles based on, for each user in the second group, i) the viewing activity of a particular user in the second group of users, ii) the viewer scores for the first group of users, and iii) a weighting value that indicates a closeness of a relationship between the particular user in the second group of users and a particular user in the first group of users; and

a profile detection module that is programmed to select, based on the content scores, one or more users in the second group of users in the remaining portion of the social network profiles as having at least a threshold likelihood of including the particular category of content.

13. The system of claim 12, wherein a particular viewer score for a particular user is determined based on a combination of scores for particular social network profiles that the particular user has viewed, as indicated by the viewing activity.

14. The system of claim 13, wherein the combination of scores for the particular social network profiles comprises a mean of the scores for the particular social network profiles.

15. The system of claim 13, wherein the combination of scores for the particular social network profiles comprises a root mean square of the scores for the particular social network profiles.

16. The system of claim 12, wherein a particular content score for a particular social network profile is determined based on a combination of particular viewer scores for particular users who have viewed the particular social network profile, as indicated by the viewing activity.

17. The system of claim 16, wherein the combination of particular viewer scores for the particular users comprises a mean of the particular viewer scores for the particular users.

18. The system of claim 16, wherein the score propagation module is further programmed to determine the weighting value for the particular users based on a number of the particular users with viewing scores that are greater than or equal to a threshold score;

and

wherein the combination of particular viewer scores for the particular users comprises a mean of the particular viewer scores for the particular users that is weighted by the weighting value for the particular users.

19. The system of claim 18, wherein the weighting value comprises a log of the number of particular users plus one.

20. A computer program product embodied in a computer readable storage device storing instructions that, when executed, cause one or more computing devices to perform operations comprising:

obtaining information that identifies viewing activity of users with regard to content that is associated with social network profiles, wherein the viewing activity identifies content that the users have viewed;

identifying a portion of the social network profiles, associated with a first group of users, that include a particular category of content;

assigning, to the portion of the social network profiles, initial scores that indicate that the portion of the social network profiles include the particular type of content;

determining viewer scores for the first group of users based on i) the viewing activity of the first group of users and ii) the initial scores that are assigned to the portion of the social network profiles;

determining content scores for each user in a second group of users in a remaining portion of the social network profiles based on, for each user in the second group, i) the viewing activity of a particular user in the second group of users, ii) the viewer scores for the first group of users, and iii) a weighting value that indicates a closeness of a relationship between the particular user in the second group of users and a particular user in the first group of users; and

selecting, based on the content scores, one or more users in the second group of users in the remaining portion of the social network profiles as having at least a threshold likelihood of including the particular category of content.