Content presentation and management system associating base content and relevant additional content

• US 20050154746A1
• Filed: 04/21/2004
• Published: 07/14/2005
• Est. Priority Date: 01/09/2004
• Status: Active Grant

• Current Assignee: R2 Solutions LLC (Acacia Research Corporation)
• Original Assignee: Yahoo! Inc. (Apollo Global Management, Inc.)
• Inventors: Yin, Hongfeng, Liu, Hongche, Mohazzab, Majid, Madhavan, Anand, Leu, Eric
• Granted Patent: US 7,483,891 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3326 using relevance feedback from the user, e.g. relevance feedback on documents, documents sets, document terms or passages
• CPC Class Codes: G06F 16/951 Indexing; Web crawling techniques
• CPC Class Codes: Y10S 707/99935 Query augmenting and refining, e.g. inexact access
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99937 Sorting

Abstract

A system for determining associations between base content and relevant content and for publishing the base content and relevant content on a client browser. The system includes a parsing module configured to parse the base content to form parsed base content; a unit-dictionary module including a plurality of query units; a unit-extraction module configured to extract query units included in the unit dictionary from the parsed base content, wherein the query units extracted from the base content are referred to as the extracted-query units; a unit-ranking module configured to rank the extracted-query units based on at least one of relevance scores and revenue-generation amounts for the query units; and a unit-matching module configured to generate at least one association between the base content and the relevant content based on whether the extracted-query units and at least one of query units and keywords associated with the relevant content are the same.

Claims

1. A system configured to associate base content with relevant content, the system comprising:

a parsing module configured to parse the base content to form parsed base content;

a unit-dictionary module including a plurality of query units;

a unit-extraction module configured to extract query units included in the unit dictionary from the parsed base content, wherein the query units extracted from the base content are referred to as the extracted-query units;

a unit-ranking module configured to rank the extracted-query units based on at least one of relevance scores and revenue-generation amounts for the query units; and

a unit-matching module configured to generate at least one association between the base content and the relevant content based on whether the extracted-query units and at least one of query units and keywords associated with the relevant content are the same.

2. The system of claim 1, further comprising, a memory configured to cache the association.

3. The system of claim 2, wherein the unit-matching module is configured to form the association in real time when a user requests base content that does not have associations cached in the memory.

4. The system of claim 3, further comprising control logic configured to cache associations formed in real time into the memory.

5. The system of claim 1, wherein the unit-matching module is configured to generate the association between the base content and the relevant content based on whether the extracted-query units and at least one of the query units and the keywords associated with the relevant content are the same or are in a same set of related query units and keywords.

6. The system of claim 1, wherein the relevant content and the base content are served to a user based on the association, if the user requests the base content.

7. The system of claim 1, wherein the relevant content includes advertisements.

8. The system of claim 1, wherein the relevant content is pointed at by a set of links that are served with the base content.

9. The system of claim 8, wherein the set of links includes at least one of a sponsor link, an integrated link, and an inside link.

10. The system of claim 8, wherein the set of links include uniform resource locators (URLs).

11. The system of claim 1, wherein the relevant content and the base content are served on a web page of a client browser.

12. The system of claim 1, wherein the system is configured to receive a uniform resource locator (URL) from a property server, and wherein the URL points at the base content on the property server.

13. The system of claim 12, wherein the parsing module is configured to retrieve the base content based on the URL.

14. The system of claim 1, further comprising a property server configured to provide a configuration file to the parsing module, and wherein the configuration file includes a set of parsing rules for parsing base content.

15. The system of claim 14, wherein the configuration file includes weighting factors for the relevance scores and the revenue-generation amounts.

16. The system of claim 14, wherein the parsing rules provide for parsing at least one of a title, metadata keywords, a metadata description, anchor text, body text, proper nouns, formatted text.

17. The system of claim 16, wherein the formatted text includes bold text and italicized text.

18. The system of claim 16, wherein the parsing rules further provide for parsing text characterized as important text by an indication in the base content.

19. The system of claim 1, wherein the unit-ranking module is configured to calculate overall-relevance scores for the extracted-query units based on the revenue-generation amounts and the relevance scores.

20. The system of claim 19, wherein the unit-matching module is configured to compare an extracted-query unit having a highest overall-relevance score with at least one of the query units and the keywords of the relevant content.

21. The system of claim 19, wherein the unit-matching module is configured to compare the extracted-query units, having overall-relevance scores greater than or equal to a threshold score, with at least one of the query units and the keywords of the relevant content.

22. The system of claim 19, wherein the overall-relevance scores are generated according to the formula:

ORS=W*RGA+(1−

W)*RS,

wherein ORS is an overall-relevance score, W is a weighting factor, RGA is a revenue-generation amount, and RS is a relevance score.

23. The system according to claim 22, wherein weighting factors are different for each property server configured to provide base content to the system.

24. A system configured to associate base content with relevant content and serve the base content with the relevant content to a user system, the system comprising:

a plurality of property servers configured to provide the base content;

a content-management system configured to receive the base content from the property servers and generate associations between the base content and the relevant content, wherein the content-management system includes;

a parsing module configured to parse the base content to form parsed base content;

a unit-extraction module configured to extract query units from the parsed based content, wherein the units extracted from the base content are referred to as the extracted-query units;

a unit-ranking module configured to rank the query units based on at least one of relevance scores and revenue-generation amounts for the query units; and

a unit-matching module configured to generate at least one association between the base content and the relevant content based on whether the extracted-query units and at least one of query units or keywords associated with the relevant content are the same; and

a client browser configured to receive the base content and the relevant content and present the base content and relevant content as a published web page.

25. The system of claim 24, further comprising, a memory configured to cache the associations.

26. The system of claim 25, wherein the unit-matching module is configured to form the associations in real time if a user requests base content that does not have associations cached in the memory.

27. The system of claim 24, wherein the relevant content and the base content are served to a user based on the association, if the user requests the base content.

28. The system of claim 24, wherein the relevant content includes advertisements.

29. The system of claim 24, wherein the relevant content is pointed at by a set of links that are served with the base content.

30. The system of claim 29, wherein the set of links includes at least one of a sponsor link, an integrated link, and an inside link.

31. The system of claim 24, wherein the property servers are configured to provide configuration files to the parsing module, and wherein the configuration files include a set of parsing rules for parsing base content.

32. The system of claim 31, wherein the parsing rules provide for parsing at least one of a title, metadata keywords, a metadata description, anchor text, body text, proper nouns, formatted text.

33. The system of claim 31, wherein the formatted text includes bold text and italicized text.

34. The system of claim 31, wherein the parsing rules further provide for parsing text characterized as important text by an indication in the base content.

35. The system of claim 24, wherein the unit-ranking module is configured to calculate overall-relevance scores for the extracted-query units based on the revenue-generation amounts and the relevance scores.

36. The system of claim 35, wherein the unit-matching module is configured to compare a extracted-query unit, having a highest overall-relevance score, with at least one of the query units and the keywords of the relevant content.

37. The system of claim 35, wherein the unit-matching module is configured to compare the extracted-query units, having overall-relevance scores greater than or equal to a threshold score, with at least one of the query units and the keywords of the relevant content.

38. The system of claim 35, wherein the overall-relevance scores are generated according to the formula:

ORS=W*RGA+(1−

W)*RS,

wherein ORS is an overall-relevance score, W is a weighting factor, RGA is a revenue-generation amount, and RS is a relevance score.

39. A method for serving base content and content that is relevant to the base content, the content being referred to as the relevant content, the method comprising:

parsing the base content to form parsed base content;

extracting a set of query units from the parsed base content based on query units in a unit-dictionary module, wherein the query units extracted from the base content are referred to as the extracted-query units;

ranking the extracted-query units based on at least one of relevance scores and revenue-generation amounts for the query units extracted from the base content;

comparing the extracted-query units with at least one of query units and keywords for the relevant content to determine matches between the extracted-query units and at least one of the query units and the keywords for the relevant content;

forming at least one association between the base content and the relevant content if the extracted-query units are the same as at least one of the query units and the keywords for the relevant content; and

serving the base content with the relevant content based on the association.

40. The method of claim 39, further comprising caching the association in a memory.

41. The method of claim 39, further comprising retrieving the association from memory to serve the base content with the relevant content if a user requests the base content.

42. The method of claim 39, further comprising serving the base content with the relevant content if the user requests the base content.

43. The system of claim 39, wherein the step of forming the association includes forming the association in real time if a user requests base content that does not have an association cached in a memory.

44. The system of claim 39, wherein the step of forming the association includes forming the association if the extracted-query units and the query units or the keywords associated the relevant content are the same or are in a same set of related query units and keywords.

45. The system of claim 39, wherein the relevant content includes advertisements.

46. The system of claim 39, wherein the relevant content is pointed at by a set of links that are served with the base content.

47. The system of claim 46, wherein the set of links includes at least one of a sponsor link, an integrated link, and an inside link.

48. The system of claim 39, wherein the ranking step includes calculating overall-relevance scores for the extracted-query units based on the revenue-generation amounts and the relevance scores.

49. The system of claim 48, wherein the comparing step includes comparing an extracted-query unit, having a highest overall-relevance score, with at least one of the query units and the keywords associated with the relevant content.

50. The system of claim 48, wherein the overall-relevance scores are generated according to the formula:

ORS=W*RGA+(1−

W)*RS,

wherein ORS is an overall-relevance score, W is a weighting factor, RGA is a revenue-generation amount, and RS is a relevance score.

51. A method for identifying related units related to other units comprising:

receiving a first unit and a second unit;

ranking related units in a related-unit dictionary that are associated with the first unit based at least one of revenue-generation amounts of the related units, a search frequency of the related units, and an identifier of a parent category of the first unit;

if at least one of the related units includes the second unit, using these related unit for relevant content identification; and

if the related units do not include the second unit, using a set of highest ranking related units for relevant content identification.

52. The method of claim 51, wherein the using steps include transferring related units to a unit-matching module for relevant content identification.

53. The method of claim 51, wherein the receiving step includes receiving the first unit and second unit in a related-unit-identification module.

54. The method of claim 53, further including extracting at least one of the first unit and the second unit from a piece of base content prior to transferring these units to the related-unit-identification module.

55. The method of claim 51, further comprising deriving at least one of the first unit and the second unit from a category.

56. The method of claim 55, wherein the category is associated with personal content.

57. The method of claim 56, wherein the personal content includes at least one of e-mail, chat room discussion, group discussion, and message-board content.

58. The method of claim 51, wherein the identifier includes a unit that identifies the parent category.

59. A context censor method comprising:

requesting a piece of base content;

classifying the piece of base content as sensitive or non-sensitive;

using any unit extracted from the base content to identify relevant content for serving with the base content, if the base content is classified as non-sensitive;

using non-offensive units extracted from the base content for identifying relevant content, if the base content is classified as sensitive;

using offensive units extracted from the base content for identifying relevant content, if at least one of an advertiser, sponsor, and user agree;

serving the base content without relevant content if the keywords are highly offensive.

60. The method of claim 59, wherein classifying the base content as sensitive or non-sensitive includes summing scores for offensive units included in the piece of base content, and comparing the sum of the scores against a threshold score.

61. The method of claim 59, and further comprising classifying units as offensive units, non-offensive units, or neutral units based on offensiveness scores of categories associated with the units.

62. The method of claim 61, and further comprising classifying the categories as offensive categories, non-offensive categories, or neutral categories based on a group of test readers offensiveness responses to the categories.