Conceptual tagging with conceptual message matching system and method

• US 8,645,379 B2
• Filed: 10/04/2010
• Issued: 02/04/2014
• Est. Priority Date: 04/27/2006
• Status: Active Grant

• Current Assignee: Seekr Technologies Incorporated
• Original Assignee: Vertical Search Works, Inc. (NTENT, Inc.)
• Inventors: Burnand, Gerald, Pool, Michael, Clark, Rob J., Matthews, Gavin
• Primary Examiner(s): Corrielus, Jean M
• Application Number:
  US12/897,649
• Publication Number:
  US 20120011120A1
• Time in Patent Office: 1,219 Days
• Field of Search: 707/732, 707/733, 707/734, 707/706, 707/749, 707/737, 707/738, 707/822, 715/760, 715/781, 709/219
• US Class Current: 707/737
• CPC Class Codes: G06Q 30/02 Marketing; Price estimation or determination; Fundraising
• CPC Class Codes: G06Q 30/0251 Targeted advertisements

Abstract

A conceptual tagging and message matching system and method are provided. In one example, the system and method generate web pages or third party web pages with pieces of content combined with the message.

Claims

1. A system for providing messages based on concepts in a piece of content, the system comprising:

a computer having a processing unit that executes that includes an indexer unit that receives one of a user requested piece of content and a message and extracts one or more features from one of the user requested piece of content and the message using an ontology,

wherein the ontology comprises a set of synsets arranged in one or more taxonomies, a semantic network linking the synsets with typed relations, one or more facets having an identifier and classification rules to determine whether a document belongs to the facet and a declarative entity list having an entity extraction rule and a function to generate a corresponding identifier;

a tag repository having a plurality of tags, wherein each tag is associated with one of the piece of content and the message, wherein each tag is determined from the one or more features extracted by the indexer unit; and

the processing unit of the computer executing including a tag manager unit that receives a user requested piece of content, matches one or more messages to the user requested piece of content based on the plurality of tags in the tag repository, selects one or more messages from the one or more matched messages and delivers the one or more selected messages with the piece of content to a user.

2. The system of claim 1, wherein the indexer unit uses an ontology to extract features from one of the user requested piece of content and the message and wherein the ontology further comprises a set of synsets arranged in one or more taxonomies and a semantic network linking the synsets with typed relations.

3. The system of claim 1, wherein the hook further comprises one or more of text, static image, animated image, embedded video, embedded audio and embedded interactive resource.

4. The system of claim 2, wherein the one or more extracted features are one or more of words, phrases, concepts, dynamically extracted entities, and document classification facets from one of the user requested piece of content and the message.

5. The system of claim 2, wherein the one or more features extracted by the indexer unit are stored in the features repository as the plurality of tags.

6. The system of claim 1, wherein the tag manager unit further comprises a tagger unit that receives the one or more features extracted from one of the piece of content and the message and supplies the tag repository with a set of tags that are derived from the one or more extracted features.

7. The system of claim 6, wherein the tagger unit distinguishes pieces of content from messages and use this information during tag determination.

8. The system of claim 1, wherein the tag manager unit matches one or more messages to the user requested piece of content based on the similarity of the plurality of tags and selects one or more messages based on the similarity of the plurality of tags.

9. The system of claim 1, wherein the tag manager unit identifies that a tag has become stale.

10. The system of claim 9 further comprising a tagging queue wherein the tag manager unit sends a request to the tagging queue to generate one or more new tags for one of the user requested piece of content and the message.

11. The system of claim 10, wherein the tag manager unit sends a request to the tagging queue when one or more tags associated with one of the user requested piece of content and the message have become stale.

12. The system of claim 10, wherein the tag manager unit sends a request to the tagging queue when there are no tags associated with one of the user requested piece of content and the message.

13. The system of claim 1, wherein the tag manager unit selects one or more default messages for the user requested piece of content when the tag manager unit is unable to match sufficient messages.

14. The system of claim 1, wherein each message comprises a hook suitable for display alongside a piece of content and a resource link to a resource so that the resource is displayed when the resource link is selected by the user.

15. The system of claim 14, wherein the ontology further comprises one or more facets comprising an identifier and classification rules that determine whether a document belongs to the facet.

16. The system of claim 14, wherein the ontology further comprises a declarative entity list comprising an entity extraction rule and a function to generate a corresponding identifier.

17. The system of claim 1, wherein each message is an advertisement.

18. The system of claim 1, wherein the user requested piece of content is a piece of content on a webpage.

19. The system of claim 1, wherein each message is an advertisement and the user requested piece of content is a piece of content on a webpage.

20. The system of claim 1, wherein the message is a related article to the user requested piece of content.

21. The system of claim 1 further comprising an indexer repository that stores the extracted features.

22. The system of claim 1, wherein the extracted features are one or more of words, phrases, concepts, dynamically extracted entities, and document classification facets from one of the user requested piece of content and the message, and wherein the tag managing unit further comprises a tagger unit that receives the features extracted from one of the piece of content and the message and supplies the tag repository with a set of tags that are derived from the extracted features.

23. The system of claim 22, wherein the tag manager unit matches one or more messages to the user requested piece of content based on the similarity of a set of tags and selects one or more messages based on the similarity of a set of tags.

24. The system of claim 23, wherein each message further comprises an advertisement and the user requested piece of content is a piece of content on a webpage.

25. A method for providing messages based on concepts in a piece of content using an indexer unit and a tag manager unit, the method comprising:

extracting, using the indexer unit, one or more features from one of a user requested piece of content and a message using ontology,

wherein the ontology comprises a set of synsets arranged in one or more taxonomies, a semantic network linking the synsets with typed relations, one or more facets having an identifier and classification rules to determine whether a document belongs to the facet and a declarative entity list having an entity extraction rule and a function to generate a corresponding identifier;

providing, in a tag repository, a plurality of tags, wherein each tag is associated with one of the piece of content and a message and each tag is determined from the extracted features;

receiving, by the tag manager unit, a user requested piece of content;

matching, using tag manager unit, one or more messages to the user requested piece of content based on the plurality of tags in the tag repository;

selecting, using tag manager unit, one or more messages from the one or more matched messages; and

delivering the one or more selected messages with the piece of content to a user.

26. The method of claim 25, wherein each message further comprises a hook suitable for display alongside a content and a resource link to a resource and wherein the method further comprising selecting, by a user, the resource link so that the resource is displayed to the user.

27. The method of claim 26, wherein the hook further comprises one or more of text, static image, animated image, embedded video, embedded audio and embedded interactive resource.

28. The method of claim 25, wherein the extracted features are one or more of words, phrases, concepts, dynamically extracted entities, and document classification facets from one of the user requested piece of content and the message.

29. The method of claim 28, wherein the extracted features are one or more of words, phrases, concepts, dynamically extracted entities, and document classification facets from one of the user requested piece of content and the message, and wherein the tag managing unit further comprises a tagger unit that receives the features extracted from one of the piece of content and the message and supplies the tag repository with a set of tags that are derived from the extracted features.

30. The system of claim 29, wherein the tag manager unit matches one or more messages to the user requested piece of content based on the similarity of a set of tags and selects one or more messages based on the similarity of a set of tags.

31. The system of claim 30, wherein each message further comprises an advertisement and the user requested piece of content is a piece of content on a webpage.

32. The method of claim 25 further comprising storing the features extracted by the indexer unit in the tag repository as the plurality of tags.

33. The method of claim 25 further comprising supplying, by a tagger unit, the tag repository with a set of tags that are derived from the extracted features.

34. The method of claim 33, wherein the tagger unit can distinguish content from messages and use this information during tag determination.

35. The method of claim 25, wherein matching the one or more messages comprising matching, using the tag manager unit, one or more messages to the user requested piece of content based on the similarity of a set of tags and wherein selecting the one or more messages further comprises selecting the one or more messages based on the similarity of a set of tags.

36. The method of claim 25 further comprising identifying, using the tag manager unit, a tag that has become stale.

37. The method of claim 36 further comprising sending a request to a tagging queue to generate one or more new tags for one of the user requested piece of content and the message.

38. The method of claim 37, wherein sending the request further comprises sending a request to the tagging queue when one or more tags associated with one of the user requested piece of content and the message have become stale.

39. The method of claim 38, wherein sending the request further comprises sending a request to the tagging queue when there are no tags associated with one of the user requested piece of content and the message.

40. The method of claim 25, wherein selecting the one or more messages comprises selecting one or more default messages for the user requested piece of content when the tag manage unit is unable to match sufficient messages.

41. The method of claim 25, wherein the ontology further comprises a set of synsets arranged in one or more taxonomies and a semantic network linking the synsets with typed relations.

42. The method of claim 25, wherein the ontology further comprises one or more facets comprising an identifier and classification rules that determine whether a document belongs to the facet.

43. The method of claim 25, wherein the ontology further comprises a declarative entity list comprising an entity extraction rule and a function to generate a corresponding identifier.

44. The method of claim 25, wherein each message further comprises an advertisement.

45. The method of claim 25, wherein the user requested piece of content is a piece of content on a webpage.

46. The method of claim 25, wherein each message further comprises an advertisement and the user requested piece of content is a piece of content on a webpage.

47. The method of claim 25, wherein the message further comprises a related article to the user requested piece of content.

48. The method of claim 25 further comprising storing the extracted features in an indexer repository.