Method and apparatus for targeting messages to users in a social network

1. A method for sending a message in a social network, the method comprising:

receiving from a source of the message or from a user a keyword relating to a subject of the message;

retrieving from an ontology source a normalized set of concepts associated with the keyword, wherein the ontology source comprises at least one community-generated ontology source;

collecting information from the social network, the information comprising one or more interactions between users of the social network and profile information for the users;

identifying a subset of the information that corresponds to the normalized set of concepts, wherein the subset of the information is used to form a global topic model comprising at least one topic;

sending the message to at least one of the users of the social network who is associated with the at least one topic; and

dynamically updating the global topic model as new information is collected from the social network,

wherein at least one of: the retrieving, the collecting, the identifying, or the dynamically updating is performed using a processor.

2. The method of claim 1 , wherein the identifying comprises:

iteratively defining the global topic model.

3. The method of claim 2 , further comprising:

assuming an initial topic model comprising at least one initial global topic;

grouping the users into one or more groups, in accordance with the at least one initial global topic;

inferring a topic for each of the one or more groups; and

generating at least one new global topic based on one or more topics inferred for the one or more groups .

4. The method of claim 3 , further comprising:

merging one or more similar topics in a set of topics comprising the at least one new global topic.

5. The method of claim 4 , wherein the merging results in each of the users being assigned to a single topic in a set of topics comprising the at least one new global topic.

6. The method of claim 4 , wherein the merging results in at least one of the users being assigned to a plurality of topics in a set of topics comprising the at least one new global topic.

7. The method of claim 1 , wherein the sending comprises:

identifying, for the at least one topic, a connection among those of the users who are associated with the at least one topic.

8. The method of claim 7 , wherein the identifying the connection comprises:

filtering the one or more interactions according to the at least one topic.

9. The method of claim 1 , wherein the one or more interactions comprise at least one of: a piece of electronic mail, an instant message, a posting to a website, a posting to a blog, a comment made on a website, a comment made on a blog, a tag made on a website, a tag made on a blog, or an online forum discussion posting.

10. The method of claim 1 , wherein the profile information comprises at least one of: data posted by one or more of the users, data provided by one or more of the users as part of a registration process, or data collected from other sources.

11. The method of claim 1 , wherein the collecting is performed in accordance with data sampling for a subset of the one or more interactions or a subset of the users.

12. The method of claim 1 , wherein the normalized set of concepts biases the global topic model and is associated with at least one penalty.

13. The method of claim 1 , wherein the ontology source comprises mapped words extracted from at least one of the one or more interactions and the profile information.

14. The method of claim 1 , further comprising:

monitoring one or more performance statistics related to the message.

15. The method of claim 14 , further comprising:

reporting the one or more performance statistics.

16. A non-transitory computer readable storage medium storing an executable program for sending a message in a social network, where the program performs steps of:

receiving from a source of the message or from a user a keyword relating to a subject of the message;

retrieving from an ontology source a normalized set of concepts associated with the keyword, wherein the ontology source comprises at least one community-generated ontology source;

collecting information from the social network, the information comprising one or more interactions between users of the social network and profile information for the users;

identifying a subset of the information that corresponds to the normalized set of concepts, wherein the subset of the information is used to form a global topic model comprising at least one topic;

sending the message to at least one of the users of the social network who is associated with the at least one topic; and

dynamically updating the global topic model as new information is collected from the social network.

17. The non-transitory computer readable storage medium of claim 16 , wherein the identifying comprises:

iteratively defining the global topic model.

18. The non-transitory computer readable storage medium of claim 17 , further comprising:

assuming an initial topic model comprising at least one initial global topic;

grouping the users into one or more groups, in accordance with the at least one initial global topic;

inferring a topic for each of the one or more groups; and generating at least one new global topic based on one or more topics inferred for the one or more groups.

19. The non-transitory computer readable storage medium of claim 18 , further comprising:

merging one or more similar topics in a set of topics comprising the at least one new global topic.

20. The non-transitory computer readable storage medium of claim 19 , wherein the merging results in each of the users being assigned to a single topic in a set of topics comprising the at least one new global topic.

21. The non-transitory computer readable storage medium of claim 19 , wherein the merging results in at least one of the users being assigned to a plurality of topics in a set of topics comprising the at least one new global topic.

22. The non-transitory computer readable storage medium of claim 16 , wherein the sending comprises:

identifying, for the at least one topic, a connection among those of the users who are associated with the at least one topic.

23. The non-transitory computer readable storage medium of claim 22 , wherein the identifying the connection comprises:

filtering the one or more interactions according to the at least one topic.

24. The non-transitory computer readable storage medium of claim 16 , wherein the one or more interactions comprise at least one of: a piece of electronic mail, an instant message, a posting to a website, a posting to a blog, a comment made on a website, a comment made on a blog, a tag made on a website, a tag made on a blog, or an online forum discussion posting.

25. The non-transitory computer readable storage medium of claim 16 , wherein the profile information comprises at least one of: data posted by one or more of the users, data provided by one or more of the users as part of a registration process, or data collected from other sources.

26. The non-transitory computer readable storage medium of claim 16 , wherein the collecting is performed in accordance with data sampling for a subset of the one or more interactions or a subset of the users.

27. The non-transitory computer readable storage medium of claim 16 , wherein the normalized set of concepts biases the global topic model and is associated with at least one penalty.

28. The non-transitory computer readable storage medium of claim 16 , wherein the ontology source comprises mapped words extracted from at least one of the one or more interactions and the profile information.

29. The non-transitory computer readable storage medium of claim 16 , further comprising:

monitoring one or more performance statistics related to the message.

30. The non-transitory computer readable storage medium of claim 29 , further comprising:

reporting the one or more performance statistics.

31. Apparatus for targeting a message in a social network, the apparatus comprising:

means for receiving from a source of the message or from a user a keyword relating to a subject of the message;

means for retrieving from an ontology source a normalized set of concepts associated with the keyword, wherein the ontology source comprises at least one community-generated ontology source;

means for collecting information from the social network, the information comprising one or more interactions between users of the social network and profile information for the users;

means for identifying a subset of the information that corresponds to the normalized set of concepts, wherein the subset of the information is used to form a global topic model comprising at least one topic;

means for sending the message to at least one of the users of the social network who is associated with the at least one topic; and

means for dynamically updating the global topic model as new information is collected from the social network.

32. The method of claim 1 , wherein the ontology source describes a relationship between the keyword and the set of normalized concepts.

33. The method of claim 32 , wherein the ontology source further describes a relationship between concepts in the set of normalized concepts.

34. The method of claim 1 , wherein the set of normalized concepts represents a meaning of the keyword to a specific subset of the users of the social network.

35. The method of claim 1 , wherein the retrieving comprises:

separately analyzing the keyword using plurality of ontology sources to produce a plurality of normalized sets of concepts;

presenting the plurality of normalized sets of concepts to the source of the message; and

receiving from the source of the message a selection of the normalized sets of concepts from among the plurality of normalized sets of concepts.

36. A method for selecting recipients to whom to send a message in a communication network, the method comprising:

receiving a keyword from a source of the message or from a user, wherein the keyword relates to a subject of the message;

analyzing the keyword in accordance with an ontology source, wherein the ontology source comprises at least one community-generated ontology source, and wherein the analyzing produces a normalized set of concepts associated with the keyword;

searching a plurality of interactions between users of the communication network for interactions that relate to the normalized set of concepts, wherein the searching results in a cluster of users of the communication network who are associated with the normalized set of concepts;

sending the message to at least one user in the cluster of users; and

dynamically updating the normalized set of concepts as new information is collected from the communication network.