Detecting content on a social network using browsing patterns

• US 8,311,950 B1
• Filed: 10/01/2009
• Issued: 11/13/2012
• Est. Priority Date: 10/01/2009
• Status: Expired due to Fees

• 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 P
• Application Number:
  US12/571,822
• Time in Patent Office: 1,139 Days
• Field of Search: None
• US Class Current: 705/319
• 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:

receiving, at a first server system, information related to users of a social network site;

determining, by the first server system, a graph for at least a portion of the social network from the received information, the graph comprising a plurality of nodes connected by directional links, where each of the nodes corresponds to a user that has a profile page on the social network, and where each of the directional links represents a user of the social network, represented by a node, having viewed at least an aspect of a profile page for another user, represented by another node;

identifying first nodes from the plurality of nodes as including content associated with a particular subject of interest;

seeding the identified first nodes with first scores, wherein the first scores indicate that first profile pages that correspond to the identified first nodes were positively identified as including content associated with the particular subject of interest;

determining second scores for second nodes of the plurality of nodes, wherein the second scores indicate likelihoods that second profile pages associated with the second nodes contain content that is associated with the particular subject of interest, wherein the second scores are determined by propagating the first scores across the social graph to the second nodes based on one or more users who have viewed aspects of both i) one or more of the first profile pages for one or more of the first nodes and ii) one or more of the second profile pages for one or more of the second nodes, and wherein the one or more users correspond to one or more third nodes of the plurality of nodes that have directional links to both i) the one or more of the first nodes and ii) the one or more of the second nodes; and

providing, by the first server system, the determined second scores for the second nodes.

2. The method of claim 1, wherein the aspects of the one or more of the first profile pages and the one or more of the second profile pages that were viewed by the one or more users comprise images identified as being associated with the particular subject of interest.

3. The method of claim 1, wherein the aspects of the one or more of the first profile pages and the one or more of the second profile pages that were viewed by the one or more users comprise videos identified as being associated with the particular subject of interest.

4. The method of claim 1, wherein determining the second scores for the second nodes comprises:

identifying the third nodes from the plurality of nodes based upon the directional links indicating that users corresponding to the third nodes viewed aspects of the one or more of the first profile pages for the first nodes that are identified associated with the first nodes;

determining third scores for the third nodes by propagating the first scores from the first nodes to the third nodes;

identifying the second nodes from the plurality of nodes based upon the directional links indicating that, in addition to having viewed aspects of the one or more of the first profile pages associated with the first nodes, the users corresponding to the third nodes also viewed aspects of the one or more of the second profile pages associated with the second nodes; and

determining the second scores for the second nodes by propagating the third scores from the third nodes to the second nodes.

5. The method of claim 4, wherein propagating the first scores from the first nodes to the third nodes comprises, for each of the third nodes, computing a weighted average score for all nodes that are connected to the third node by directional links indicating that a user of the social network corresponding to the third node viewed aspects of the one or more of the first profile pages for the first nodes and aspects of the one or more of the second profile pages for the second nodes.

6. The method of claim 4, wherein propagating the third scores from the third nodes to the second nodes comprises, for each of the second nodes, computing a weighted average score for all nodes that are connected to the second node by directional links indicating that users of the social network corresponding to nodes viewed an aspect of a profile page for the second node.

7. The method of claim 1, wherein there first nodes are different than the second nodes.

8. The method of claim 1, wherein there is overlap between the first nodes and the second nodes.

9. The method of claim 1, wherein the first scores are propagated from the first nodes to the second nodes through multiple iterations of determining the second scores for the second nodes;

and

the method further comprises, for each of the second nodes, dampening the second score for the second node based upon an iteration at which a score is first propagated to the second node, wherein a score first propagated to a node at an earlier iteration is dampened less than a score first propagated to a node at a later iteration.

10. The method of claim 1, further comprising, for each of the second nodes, boosting the second score for the second node based upon a number of third nodes that are connected to the second node by directional links, wherein a score for a node having a greater number of directional links to the third nodes is boosted by a greater amount than a score for a node having a lesser number of directional links to the third nodes.

11. The method of claim 1, wherein the second scores for the second nodes are determined through multiple iterations of propagating the first scores from the first nodes to the second nodes.

12. The method of claim 11, wherein the multiple iterations comprises at least four iterations.

13. The method of claim 1, further comprising identifying a portion of the second nodes that have a second score of at least a threshold level as positively including content associated with the particular subject of interest.

14. The method of claim 1, wherein the particular subject of interest comprises at least one of the group consisting of adult content and illegal content.

15. The method of claim 1, further comprising:

receiving updated information related to the users of the social network site;

revising the graph using the updated information; and

updating scores for the plurality of nodes based upon the revised graph.

16. A system for detecting profile pages on a social network as including content associated with a particular subject of interest, the system comprising:

one or more servers;

an interface for the one or more servers that receives information related to users of a social network site;

a social graph linking module configured to determine a graph for at least a portion of the social network from the information received by the interface, the graph comprising a plurality of nodes connected by directional links, where each of the nodes corresponds to a user that has a profile page on the social network, and where each of the directional links represents a user of the social network, represented by a node, having viewed at least an aspect of a profile page for another user, represented by another node;

a score seeding component that identifies first nodes from the plurality of nodes as including content associated with a particular subject of interest and seeds the identified first nodes with first scores, wherein the first scores indicate that first profile pages that correspond to the identified first nodes were positively identified as including content associated with the particular subject of interest;

an iterative score propagation module that is configured to determine second scores for second nodes of the plurality of nodes, wherein the second scores indicate likelihoods that second profile pages associated with the second nodes contain content that is associated with the particular subject of interest, wherein the second scores are determined by propagating the first scores across the social graph to the second nodes based on one or more users who have viewed aspects of both i) one or more of the first profile pages for one or more of the first nodes and ii) one or more of the second profile pages for one or more of the second nodes, and wherein the one or more users correspond to third nodes of the plurality of nodes that have directional links to both i) the one or more of the first nodes and ii) the one or more of the second nodes; and

a content detection module that is configured to provide the determined second scores for the second nodes.

17. The system of claim 16, wherein the content detection module is further configured to identify a portion of the second nodes that have a second score of at least a threshold level as positively including content associated with the particular subject of interest.

18. The system of claim 16, wherein the first scores are propagated from the first nodes to the second nodes through multiple iterations of determining the second scores for the second nodes;

and

the system further comprising a score dampening component that is configured to, for each of the second nodes, dampen the second score for the second node based upon an iteration at which a score is first propagated to the second node, wherein a score first propagated to a node at an earlier iteration is dampened less than a score first propagated to a node at a later iteration.

19. The system of claim 16, further comprising a score boosting component configured to:

identify a group of the third nodes as being sufficiently associated with the particular subject of interest based upon directional links between the third nodes and the first nodes, and

for each of the second nodes, boost the second score for the second node based upon a number of third nodes from the group of third nodes that are connected to the second node by directional links, wherein a score for a node having a greater number of directional links to the identified group of third nodes is boosted by a greater amount than a score for a node having a lesser number of directional links to the identified group of third nodes.

20. The system of claim 16, wherein the aspects of the one or more of the first profile pages and the one or more of the second profile pages that were viewed by the one or more users comprise images identified as being associated with the particular subject of interest.

21. The system of claim 16, wherein the aspects of the one or more of the first profile pages and the one or more of the second profile pages that were viewed by the one or more users comprise videos identified as being associated with the particular subject of interest.

22. The system of claim 16, wherein the iterative score propagation module is further configured to:

identify the third nodes from the plurality of nodes based upon the directional links indicating that users corresponding to the third nodes viewed aspects of the one or more of the first profile pages for the first nodes that are identified associated with the first nodes;

determine third scores for the third nodes by propagating the first scores from the first nodes to the third nodes;

identify the second nodes from the plurality of nodes based upon the directional links indicating that, in addition to having viewed aspects of the one or more of the first profile pages associated with the first nodes, the users corresponding to the third nodes also viewed aspects of the one or more of the second profile pages associated with the second nodes;

and determine the second scores for the second nodes by propagating the third scores from the third nodes to the second nodes.

23. The system of claim 22, wherein the first scores are propagated from the first nodes to the third nodes based on, for each of the third nodes, a computation of a weighted average score for all nodes that are connected to the third node by directional links indicating that a user of the social network corresponding to the third node viewed aspects of the one or more of the first profile pages for the first nodes and aspects of the one or more of the second profile pages for the second nodes.

24. The system of claim 22, wherein the third scores are propagated from the third nodes to the second nodes based on, for each of the second nodes, a computation of a weighted average score for all nodes that are connected to the second node by directional links indicating that users of the social network corresponding to nodes viewed an aspect of a profile page for the second node.

25. The system of claim 16, wherein there first nodes are different than the second nodes.

26. The system of claim 16, wherein there is overlap between the first nodes and the second nodes.

27. The system of claim 16, wherein the second scores for the second nodes are determined by the iterative score propagation module through multiple iterations of propagating the first scores from the first nodes to the second nodes.

28. The system of claim 27, wherein the multiple iterations comprises at least four iterations.

29. The system of claim 16, wherein the particular subject of interest comprises at least one of the group consisting of adult content and illegal content.

30. The system of claim 16, wherein:

the interface is further configured to receive updated information related to the users of the social network site;

the social graph linking module is further configured to revise the graph using the updated information; and

the iterative score propagation module is further configured to update scores for the plurality of nodes based upon the revised graph.

31. A computer program product tangibly embodied on one or more computer readable media storing instructions that, when executed by one or more computing devices, cause operations to be performed comprising:

receiving information related to users of a social network site;

determining a graph for at least a portion of the social network from the received information, the graph comprising a plurality of nodes connected by directional links, where each of the nodes corresponds to a user that has a profile page on the social network, and where each of the directional links represents a user of the social network, represented by a node, having viewed at least an aspect of a profile page for another user, represented by another node;

identifying first nodes from the plurality of nodes as including content associated with a particular subject of interest;

seeding the identified first nodes with first scores, wherein the first scores indicate that first profile pages that correspond to the identified first nodes were positively identified as including content associated with the particular subject of interest;

determining second scores for second nodes of the plurality of nodes, wherein the second scores indicate likelihoods that second profile pages associated with the second nodes contain content that is associated with the particular subject of interest, wherein the second scores are determined by propagating the first scores across the social graph to the second nodes based on one or more users who have viewed aspects of both i) one or more of the first profile pages for one or more of the first nodes and ii) one or more of the second profile pages for one or more of the second nodes, and wherein the one or more users correspond to one or more third nodes of the plurality of nodes that have directional links to both i) the one or more of the first nodes and ii) the one or more of the second nodes; and

providing the determined second scores for the second nodes.

32. The computer program product of claim 31, wherein the aspects of the one or more of the first profile pages and the one or more of the second profile pages that were viewed by the one or more users comprise images identified as being associated with the particular subject of interest.

33. The computer program product of claim 31, wherein the aspects of the one or more of the first profile pages and the one or more of the second profile pages that were viewed by the one or more users comprise videos identified as being associated with the particular subject of interest.

34. The computer program product of claim 31, wherein determining the second scores for the second nodes comprises:

identifying the third nodes from the plurality of nodes based upon the directional links indicating that users corresponding to the third nodes viewed aspects of the one or more of the first profile pages for the first nodes that are identified associated with the first nodes;

determining third scores for the third nodes by propagating the first scores from the first nodes to the third nodes;

identifying the second nodes from the plurality of nodes based upon the directional links indicating that, in addition to having viewed aspects of the one or more of the first profile pages associated with the first nodes, the users corresponding to the third nodes also viewed aspects of the one or more of the second profile pages associated with the second nodes; and

determining the second scores for the second nodes by propagating the third scores from the third nodes to the second nodes.

35. The computer program product of claim 34, wherein propagating the first scores from the first nodes to the third nodes comprises, for each of the third nodes, computing a weighted average score for all nodes that are connected to the third node by directional links indicating that a user of the social network corresponding to the third node viewed aspects of the one or more of the first profile pages for the first nodes and aspects of the one or more of the second profile pages for the second nodes.

36. The computer program product of claim 34, wherein propagating the third scores from the third nodes to the second nodes comprises, for each of the second nodes, computing a weighted average score for all nodes that are connected to the second node by directional links indicating that users of the social network corresponding to nodes viewed an aspect of a profile page for the second node.

37. The computer program product of claim 31, wherein there first nodes are different than the second nodes.

38. The computer program product of claim 31, wherein there is overlap between the first nodes and the second nodes.

39. The computer program product of claim 31, wherein the first scores are propagated from the first nodes to the second nodes through multiple iterations of determining the second scores for the second nodes;

and

the operations further comprising, for each of the second nodes, dampening the second score for the second node based upon an iteration at which a score is first propagated to the second node, wherein a score first propagated to a node at an earlier iteration is dampened less than a score first propagated to a node at a later iteration.

40. The computer program product of claim 31, the operations further comprising, for each of the second nodes, boosting the second score for the second node based upon a number of third nodes that are connected to the second node by directional links, wherein a score for a node having a greater number of directional links to the third nodes is boosted by a greater amount than a score for a node having a lesser number of directional links to the third nodes.

41. The computer program product of claim 31, wherein the second scores for the second nodes are determined through multiple iterations of propagating the first scores from the first nodes to the second nodes.

42. The computer program product of claim 41, wherein the multiple iterations comprises at least four iterations.

43. The computer program product of claim 31, the operations further comprising identifying a portion of the second nodes that have a second score of at least a threshold level as positively including content associated with the particular subject of interest.

44. The computer program product of claim 31, wherein the particular subject of interest comprises at least one of the group consisting of adult content and illegal content.

45. The computer program product of claim 31, the operations further comprising:

receiving updated information related to the users of the social network site;

revising the graph using the updated information; and

updating scores for the plurality of nodes based upon the revised graph.