Detecting associates

1. A method of detecting preference events associated with a content item, the method comprising:

receiving, for the content item, a plurality of preference events from a plurality of users, wherein each of multiple types of preference events was initiated with a separate control;

detecting associated events within the received preference events, the associated events comprising multiple preference events of the same type;

identifying a cohort of users from whom the associated events were received,

wherein a group of users within the plurality of users constitutes a cohort if a first threshold percentage of users in the group of users submit the same type of preference event regarding a second threshold percentage of activities of one or more other users in the group of users;

dynamically adjusting the cohort as additional associated events are received from additional users; and

determining whether to promote the content item, wherein said determining includes reducing the effect of associated events received from the cohort;

wherein promoting the content item comprises increasing the prominence of the content item in an interface.

2. The method of claim 1 wherein a type of preference event is one of a digg, a bury, and a comment.

3. The method of claim 1 wherein a first type of preference event is an indication that the content should be promoted.

4. The method of claim 1 wherein reducing the effect of the events includes increasing a score associated with the content.

5. The method of claim 1 wherein reducing the effect of the events includes decreasing a score associated with the content.

6. The method of claim 1 wherein identifying a cohort includes determining a group of associated users.

7. The method of claim 6 further comprising determining whether the same preference event was received from a threshold number of the associated users.

8. The method of claim 6 wherein the group is stored in a data table configured to identify the group of associated users and no other users of known associates.

9. The method of claim 8 further comprising periodically updating the table.

10. The method of claim 1 wherein detecting associated events includes determining whether more than one of the associated events were received from a single Internet Protocol (IP) address.

11. The method of claim 1 wherein reducing the effect includes discarding the associated events when determining whether to promote the item.

12. The method of claim 1 wherein reducing the effect includes weighting the associated events when determining whether to promote the item.

13. A system for detecting, for a content item, associated preference events associated with a content item, including:

a processor; and

a memory coupled with the processor, wherein the memory is configured to provide the processor with instructions which when executed cause the processor to:

receive, for the content item, a plurality of preference events from a plurality of users, wherein each of multiple types of preference events was initiated with a separate control;

detect associated events within the received preference events, the associated events comprising multiple preference events of the same type;

identify a cohort of users from whom the associated events were received, wherein a group of users within the plurality of users constitutes a cohort if a first threshold percentage of users in the group of users submit the same type of preference event regarding a second threshold percentage of activities of one or more other users in the group of users;

dynamically adjust the cohort as additional associated events are received from additional users; and

determine whether to promote the content item, wherein said determining includes reducing the effect of associated events received from the cohort;

wherein promoting the content item comprises increasing the prominence of the content item in an interface.

14. The system of claim 13 wherein a type of preference event is one of a digg, a bury, and a comment.

15. The system of claim 13 wherein a first type of preference event is an indication that the content should be promoted.

16. The system of claim 13 wherein identifying a cohort includes determining a group of associated users.

17. The system of claim 13 wherein reducing the effect includes discarding the associated events when determining whether to promote the item.

18. The system of claim 13 wherein reducing the effect includes weighting the associated events when determining whether to promote the item.

19. A computer program product for detecting preference events associated with a content item, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:

receiving, for the content item, a plurality of preference events from a plurality of users, wherein each of multiple types of preference events is initiated with a separate control;

detecting associated events within the received preference events, the associated events comprising multiple preference events of the same type;

identifying a cohort of users from whom the associated events were received, wherein a group of users within the plurality of users constitutes a cohort if a first threshold percentage of users in the group of users submit the same type of preference event regarding a second threshold percentage of activities of one or more other users in the group of users;

dynamically adjusting the cohort as additional associated events are received from additional users; and

determining whether to promote the content item, wherein said determining includes reducing the effect of associated events received from the cohort;

wherein promoting the content item comprises increasing the prominence of the content item in an interface.

20. The method of claim 1 , wherein identifying a cohort of users comprises, for each user in the plurality of users:

correlating the user's activity regarding the content item with activity regarding the content item by each other user in the plurality of users;

wherein every user in the plurality of users regularly visits a same web site and submits preference events regarding activity at the same web site by other users in the plurality of users.

21. The method of claim 1 , wherein a set of users within the plurality of users constitutes a cohort if a first threshold percentage of users in the set of users submit the same type of preference event regarding a second threshold percentage of activities of other users in the set of users.

22. The method of claim 1 , wherein identifying a cohort of users comprises, for each user in the plurality of users:

identifying a preference event submitted by the user for a given content item; and

determining whether a threshold percentage of other users in the plurality of users have submitted the same type of preference event for the given content item.

23. The method of claim 1 , wherein identifying a cohort of users comprises:

constructing a data table in which:

each cell in a first dimension corresponds to one of the plurality of users; and

each cell in a second dimension corresponds to one of a plurality of content items;

for each cell corresponding to a given user and to a content item regarding which the corresponding user submitted a preference event:

populating the cell with identities of users, other than the given user, who submitted the same type of preference event as the given user; and

determining whether a threshold number or a threshold percentage of cells corresponding to the given user include a common set of identities of users other than the given user;

wherein if the threshold number or the threshold percentage of cells corresponding to the given user include a common set of identities of users other than the given user, the given user and the users other than the given user constitute a cohort.