Identifying items that have experienced recent interest bursts

• US 20080155465A1
• Filed: 12/21/2006
• Published: 06/26/2008
• Est. Priority Date: 12/21/2006
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Chi, Liang-Yu (Tom), Xu, Zhichen, Fu, Yun
• Granted Patent: US 8,046,248 B2
• Time in Patent Office: Days
• US Class Current: 715/803
• CPC Class Codes: G06F 16/9535 Search customisation based on user profiles and personalisation
• CPC Class Codes: G06Q 10/063 Operations research, analysis or management

Abstract

Techniques are described for identifying items that have recently undergone an interest burst. Items that have recently undergone an interest burst are identified by comparing how many interest-actions have been performed on the items during a current time window against how many interest-actions have been performed on the items historically. Various tests are performed to rule out candidates that are not likely to be of interest to other users. In addition, various spam detection techniques are described for reducing the possibility that the items that are listed as interest burst items are listed because of spam.

Claims

1. A method for identifying items that have experienced interest bursts, comprising the computer-implemented steps of:

for each item of a plurality of items, performing the steps of

during a particular window of time, maintaining an interest value based on how many interest-actions are performed on said item during said particular window of time; and

determining whether the item has experienced an interest burst based, at least in part, on whether the interest value is greater than a particular fraction of a historical indication of interest for the item.

2. The method of claim 1 wherein the plurality of items are tags.

3. The method of claim 2 wherein the interest value for a given tag is based, at least in part, on reputation values associated with users that performed interest-actions on the given tag during the particular window of time.

4. The method of claim 2 wherein:

the interest value for a given tag reflects a particular interest-action in which a particular user assigned the given tag with a document during the particular window of time, and

the method includes assigning to said particular interest-action a weight that is based, at least in part, on how many other tags the particular user assigned to said particular document.

5. The method of claim 1 wherein:

the particular window of time is a first window;

the step of determining whether the item has experienced an interest burst based, at least in part, on whether the interest value is greater than a particular fraction of a historical indication of interest for the item, is performed prior to the end of the first window;

the method further comprises the steps of;

during the first window, maintaining a second interest value based on how many interest-actions were performed on said item during a second time window that begins during the first window; and

after the end of the first window, determining whether the item has experienced an interest burst based, at least in part, on whether the second interest value is greater than a particular fraction of a historical indication of interest for the item.

6. The method of claim 1 wherein the step of determining whether the item has experienced an interest burst is also based, at least in part, on whether the interest value is greater than a predetermined minimum threshold.

7. The method of claim 1 wherein the step of determining whether the item has experienced an interest burst is also based, at least in part, on whether the interest value is less than a predetermined maximum threshold.

8. The method of claim 1 wherein:

the interest-actions include private interest-actions and public interest-actions; and

the step of determining whether the item has experienced an interest burst is also based, at least in part, on whether the number of public interest-actions is greater than a predetermined threshold.

9. The method of claim 1 further comprising:

ranking those items that have experienced an interest burst;

generating a list that includes the highest-ranked items that have experienced an interest burst; and

providing the list for display to a user over a network.

10. The method of claim 9 wherein:

the step of ranking those items that have experienced an interest burst is based on scores generated for the items;

the score for each items is generated based on

the interest value for the item;

the historical indication for the item; and

a number of windows that preceded the particular window.

20. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 1.

21. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 2.

22. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 3.

23. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 4.

24. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 5.

25. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 6.

26. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 7.

27. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 8.

28. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 9.

29. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 10.

11. A method for identifying items that have experienced interest bursts, comprising the computer-implemented steps of:

concurrently maintaining a plurality of count sets, wherein each count set indicates interest-actions that are performed on items during a particular time window of a plurality of time windows;

determining which of said time windows is a current window, wherein the current window is an active window with an earliest start time of the plurality of time windows;

determining which items have experienced interest bursts based on values in the count set associated with the current window;

at the end of the current window, determining which of said time windows is a new current window, wherein the new current window is an active window with an earliest start time of the plurality of time windows other than the current window; and

after the current window has ended, determining which items have experienced interest bursts based on values in the count set associated with the new current window.

12. The method of claim 11 further comprising, in response to the end of the current window:

starting a new window;

resetting the values in the count set associated with the current window; and

using the count set associated with the new window to begin counting interest-actions for the new window.

30. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 11.

31. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 12.

13. A method for identifying items that have experienced interest bursts, comprising the computer-implemented steps of:

identifying a set of interest burst item candidates;

wherein identifying a set of interest burst item candidates includes, for each item of a plurality of items, performing the steps of

during a particular window of time, maintaining an interest value based on how many interest-actions are performed on said item during said particular window of time; and

determining whether the item has experienced an interest burst based, at least in part, on the interest value and a historical indication of interest for the item;

for each item in the set of interest burst item candidates, determining whether the interest burst experienced by the item is likely to have been the result of spamming;

ranking the items in the set of interest burst item candidates, wherein the ranking of each item is based, at least in part, on whether the interest burst experienced by the item is likely to have been the result of spamming;

generating a list that include the highest-ranked items that have experienced an interest burst; and

providing the list for display to a user over a network.

14. The method of claim 13 wherein determining whether the interest burst experienced by the item is likely to have been the result of spamming includes performing a least one test to detect whether a particular user that performed interest actions related to the item is likely to be a spammer.

15. The method of claim 13 wherein determining whether the interest burst experienced by the item is likely to have been the result of spamming includes performing a least one test to detect whether a plurality of users that performed interest actions related to the item are likely to have cooperated to spam the item.

16. The method of claim 15 wherein performing a least one test to detect whether a plurality of users that performed interest actions related to the item are likely to have cooperated to spam the item includes performing a test based on registration times of the plurality of users.

17. The method of claim 15 wherein performing a least one test to detect whether a plurality of users that performed interest actions related to the item are likely to have cooperated to spam the item includes performing a test based on similarity between the identifiers user by the plurality of users

18. The method of claim 15 wherein performing a least one test to detect whether a plurality of users that performed interest actions related to the item are likely to have cooperated to spam the item includes performing a test based on interest-action similarity between the plurality of users.

19. The method of claim 13 wherein determining whether the interest burst experienced by the item is likely to have been the result of spamming includes performing a least one test to detect whether content of the item indicates that the item is likely to be spam.

32. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 13.

33. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 14.

34. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 15.

35. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 16.

36. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 17.

37. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 18.

38. A computer-readable medium carrying one or more sequences of instructions which, when executed by one or more processors, causes the one or more processors to perform the method recited in claim 19.