Content information analyzing method and apparatus

• US 20030120649A1
• Filed: 02/10/2003
• Published: 06/26/2003
• Est. Priority Date: 11/26/2001
• Status: Active Grant

• Current Assignee: Fujitsu Limited
• Original Assignee: Fujitsu Limited
• Inventors: Uchino, Kanji, Kume, Yuki
• Granted Patent: US 7,814,043 B2
• Time in Patent Office: Days
• US Class Current: 707/5
• CPC Class Codes: G06F 16/30 of unstructured textual data document management systems G06F16/93

Abstract

This invention is to automatically extract noteworthy information from a large amount of information. First, a disclosure unit of an individual opinion such as a statement in a personal Web page or a bulletin board is extracted from collected content information, and information such as URL or statement number for specifying the disclosure unit of the individual opinion is registered. Next, an object such as company name or industry type of the individual opinion is specified. Then, the disclosed contents of the individual opinion are analyzed, so that an evaluation as to the object such as good evaluation or bad evaluation is specified. Besides, the reliability is determined based on referenced degree ranking and based on whether information to indicate the basis of the opinion or the identity of the speaker is included. Thus, the evaluation as to the object as characteristics of the individual opinion can be presented to requesters. Besides, for example, only a bad evaluation can be extracted from evaluations as to the object of the individual opinion. Furthermore, the opinion, which has a high influence degree and is noteworthy, can also be found based on the referenced degree ranking or the reliability.

Claims

1. A content information analyzing method, comprising the steps of:

extracting a disclosure unit of an individual opinion from collected content information;

specifying an object of said individual opinion; and

analyzing a disclosed content of said individual opinion and specifying an evaluation as to said object of said individual opinion.

2. The content information analyzing method as set forth in claim 1, wherein said extracting step comprises the steps of:

specifying a unit of said collected content information including said individual opinion; and

extracting said disclosure unit of said individual opinion from the specified unit of said collected content information.

3. The content information analyzing method as set forth in claim 2, wherein said step of specifying a unit of said collected content information is carried out in descending order of a referenced degree for each said unit of said collected content information.

4. The content information analyzing method as set forth in claim 1, wherein said extracting step comprises a step of detecting a group of said disclosure units of said individual opinions by tracing a reference source of said individual opinion.

5. The content information analyzing method as set forth in claim 1, wherein said extracting step comprises a step of specifying a category as to said object of said individual opinion.

6. The content information analyzing method as set forth in claim 5, wherein said analyzing step comprises a step of analyzing a disclosed content of said individual opinion based on said category as to said object of said individual opinion and specifying an evaluation as to said object of said individual opinion.

7. The content information analyzing method as set forth in claim 1, further comprising a step of judging whether information that can be a basis of said individual opinion is included in said disclosure unit of said individual opinion.

8. The content information analyzing method as set forth in claim 1, further comprising a step of specifying a genre of said disclosed content of said individual opinion.

9. The content information analyzing method as set forth in claim 1, further comprising a step of determining reliability of said disclosure unit of said individual opinion.

10. The content information analyzing method as set forth in claim 9, wherein said determining step comprises a step of judging whether information indicating an identity of the individual is included in said disclosure unit of said individual opinion.

11. The content information analyzing method as set forth in claim 9, wherein said determining step comprises a step of judging whether information that can be a basis of said individual opinion is included in said disclosure unit of said individual opinion.

12. The content information analyzing method as set forth in claim 1, wherein said step of specifying an object of said individual opinion comprises a step of specifying an object of said individual opinion by using a dictionary on at least one of a Uniform Resource Locator (URL), a company name, an abbreviation, and an industry type.

13. The content information analyzing method as set forth in claim 12, further comprising a step of registering information concerning an industry type corresponding to a company name into said dictionary by using a URL of said collected content information and a similar URL registered in said dictionary.

14. The content information analyzing method as set forth in claim 12, further comprising a step of registering an abbreviation into said dictionary by using anchored character information on said collected content information and a link destination URL of said collected content information.

15. The content information analyzing method as set forth in claim 12, further comprising a step of registering information concerning an industry type corresponding to a company name into said dictionary by using information of a link topology obtained by analyzing a link relation among said collected content information.

16. The content information analyzing method as set forth in claim 12, further comprising a step of extracting a feature word from said collected content information, specifying an industry type based on the extracted feature word by using a feature word dictionary including feature words as to respective industry types, and registering information concerning an industry type corresponding to a company name into said dictionary.

17. The content information analyzing method as set forth in claim 5, wherein said step of specifying a category comprises a step of specifying an industry type of a company, which is an object of said individual opinion, by using a second dictionary as to feature words corresponding to respective industry types.

18. The content information analyzing method as set forth in claim 17, further comprising a step of extracting a feature word from said collected content information in which an industry type is specified, and adding the extracted feature word into said second dictionary correspondingly to said industry type.

19. The content information analyzing method as set forth in claim 17, further comprising a step of identifying, in a search log for said collected content information, a keyword of a search in a state where an industry type is already specified, and registering the identified keyword as a feature word into said second dictionary.

20. The content information analyzing method as set forth in claim 12, further comprising the steps of:

judging whether a jump destination URL of a searcher included in a search log for said collected content information is included in said dictionary; and

adding a search keyword included in said search log to said dictionary if it is judged to be included in said dictionary.

21. The content information analyzing method as set forth in claim 1, further comprising the steps of:

analyzing a plurality of search logs for the content information, each said search log including a search keyword and a jump destination, and specifying a site including a personal homepage in a lower layer based on a number of accesses and a number of kinds of search keywords; and

extracting said personal homepage included in the lower layer of the specified site as said disclosure unit of said opinion of said individual.

22. The content information analyzing method as set forth in claim 1, further comprising a step of analyzing a plurality of search logs for the content information, each said search log including a search keyword and a jump destination URL, and extracting a site of an individual or a small-scale organization based on a number of accesses and a number of kinds of search keywords as said disclosure unit of said opinion of said individual.

23. The content information analyzing method as set forth in claim 12, further comprising the steps of:

analyzing a plurality of search logs for the content information, each said search log including a search keyword and a jump destination, and specifying a site including a personal homepage in a lower layer based on a number of accesses and a number of kinds of search keywords; and

registering information concerning the specified site into said dictionary.

24. The content information analyzing method as set forth in claim 12, further comprising the steps of:

analyzing a plurality of search logs for the content information, each said search log including a search keyword and a jump destination, and specifying a company site based on a number of accesses and a number of kinds of search keywords; and

registering information concerning the specified company site into said dictionary.

25. A content information analyzing method, comprising the steps of:

extracting a disclosure unit of an individual opinion from collected content information;

specifying an object of said individual opinion; and

determining reliability of said disclosure unit of said individual opinion.

26. A content information analyzing method, comprising the steps of:

obtaining a plurality of search logs for the content information, each said search log including a search keyword and a jump destination URL;

counting a number of accesses and a number of kinds of search keywords for each said jump destination URL in accordance with a predetermined rule;

totaling said number of accesses and said number of kinds of search keywords up for the jump destination URLs included in lower layers of each site specified based on a structure of the URL; and

identifying a type of said site based on the totaled number of accesses and the totaled number of kinds of search keywords for said site.

27. The content information analyzing method as set forth in claim 26, wherein said obtaining step comprises a step of generating a plurality of log records, each said log records including a search keyword and a jump destination, from log data stored in a proxy server.

28. The content information analyzing method as set forth in claim 26, wherein said counting step comprises the steps of:

normalizing the search keyword; and

deleting a search log other than a last search log if a plurality of search logs relating the same keyword within a predetermined time are included.

29. A content information analyzing apparatus, comprising:

means for extracting a disclosure unit of an individual opinion from collected content information;

means for specifying an object of said individual opinion; and

means for analyzing a disclosed content of said individual opinion and specifying an evaluation as to said object of said individual opinion.

30. The content information analyzing apparatus as set forth in claim 29, wherein said means for extracting comprises:

means for specifying a unit of said collected content information including said individual opinion; and

means for extracting said disclosure unit of said individual opinion from the specified unit of said collected content information.

31. The content information analyzing apparatus as set forth in claim 30, wherein said means for specifying a unit of said collected content information processes said collected content information in descending order of a referenced degree for each said unit of said collected content information.

32. The content information analyzing apparatus as set forth in claim 29, wherein said means for extracting comprises means for detecting a group of said disclosure units of said individual opinions by tracing a reference source of said individual opinion.

33. The content information analyzing apparatus as set forth in claim 29, wherein said means for extracting comprises means for specifying a category as to said object of said individual opinion.

34. The content information analyzing apparatus as set forth in claim 33, wherein said means for analyzing comprises means for analyzing a disclosed content of said individual opinion based on said category as to said object of said individual opinion and specifying an evaluation as to said object of said individual opinion.

35. The content information analyzing method as set forth in claim 29, further comprising means for judging whether information that can be a basis of said individual opinion is included in said disclosure unit of said individual opinion.

36. The content information analyzing apparatus as set forth in claim 29, further comprising means for specifying a genre of said disclosed content of said individual opinion.

37. The content information analyzing apparatus as set forth in claim 29, further comprising means for determining reliability of said disclosure unit of said individual opinion.

38. The content information analyzing apparatus as set forth in claim 37, wherein said means for determining comprises means for judging whether information indicating an identity of the individual is included in said disclosure unit of said individual opinion.

39. The content information analyzing apparatus as set forth in claim 37, wherein said means for determining comprises means for judging whether information that can be a basis of said individual opinion is included in said disclosure unit of said individual opinion.

40. The content information analyzing apparatus as set forth in claim 29, wherein said means for specifying an object of said individual opinion comprises means for specifying an object of said individual opinion by using a dictionary on at least one of a Uniform Resource Locator (URL), a company name, an abbreviation, and an industry type.

41. The content information analyzing apparatus as set forth in claim 40, further comprising means for registering information concerning an industry type corresponding to a company name into said dictionary by using a URL of said collected content information and a similar URL registered in said dictionary.

42. The content information analyzing apparatus as set forth in claim 40, further comprising means for registering an abbreviation into said dictionary by using anchored character information on said collected content information and a link destination URL of said collected content information.

43. The content information analyzing apparatus as set forth in claim 40, further comprising means for registering information concerning an industry type corresponding to a company name into said dictionary by using information of a link topology obtained by analyzing a link relation among said collected content information.

44. The content information analyzing apparatus as set forth in claim 40, further comprising means for extracting a feature word from said collected content information, specifying an industry type based on the extracted feature word by using a feature word dictionary including feature words as to respective industry types, and registering information concerning an industry type corresponding to a company name into said dictionary.

45. The content information analyzing apparatus as set forth in claim 33, wherein said means for specifying a category comprises means for specifying an industry type of a company, which is an object of said individual opinion, by using a second dictionary as to feature words corresponding to respective industry types.

46. The content information analyzing apparatus as set forth in claim 45, further comprising means for extracting a feature word from said collected content information in which an industry type is specified, and adding the extracted feature word into said second dictionary correspondingly to said industry type.

47. The content information analyzing apparatus as set forth in claim 45, further comprising means for identifying, in a search log for said collected content information, a keyword of a search in a state where an industry type is already specified, and registering the identified keyword as a feature word into said second dictionary.

48. The content information analyzing apparatus as set forth in claim 39, further comprising:

means for judging whether a jump destination URL of a searcher included in a search log for said collected content information is included in said dictionary; and

means for adding a search keyword included in said search log to said dictionary if it is judged to be included in said dictionary.

49. The content information analyzing apparatus as set forth in claim 29, further comprising:

means for analyzing a plurality of search logs for the content information, each said search log including a search keyword and a jump destination, and specifying a site including a personal homepage in a lower layer based on a number of accesses and a number of kinds of search keywords; and

means for extracting said personal homepage included in the lower layer of the specified site as said disclosure unit of said opinion of said individual.

50. The content information analyzing apparatus as set forth in claim 29, further comprising means for analyzing a plurality of search logs for the content information, each said search log including a search keyword and a jump destination URL, and extracting a site of an individual or a small-scale organization based on a number of accesses and a number of kinds of search keywords as said disclosure unit of said opinion of said individual.

51. The content information analyzing apparatus as set forth in claim 40, further comprising:

means for analyzing a plurality of search logs for the content information, each said search log including a search keyword and a jump destination, and specifying a site including a personal homepage in a lower layer based on a number of accesses and a number of kinds of search keywords; and

means for registering information concerning the specified site into said dictionary.

52. The content information analyzing apparatus as set forth in claim 40, further comprising:

means for analyzing a plurality of search logs for the content information, each said search log including a search keyword and a jump destination, and specifying a company site based on a number of accesses and a number of kinds of search keywords; and

means for registering information concerning the specified company site into said dictionary.

53. A content information analyzing apparatus, comprising:

means for extracting a disclosure unit of an individual opinion from collected content information;

means for specifying an object of said individual opinion; and

means for determining reliability of said disclosure unit of said individual opinion.

54. A content information analyzing apparatus, comprising:

means for obtaining a plurality of search logs for the content information, each said search log including a search keyword and a jump destination URL;

a counter for counting a number of accesses and a number of kinds of search keywords for each said jump destination URL in accordance with a predetermined rule;

means for totaling said number of accesses and said number of kinds of search keywords up for the jump destination URLs included in lower layers of each site specified based on a structure of the URL; and

means for identifying a type of said site based on the totaled number of accesses and the totaled number of kinds of search keywords for said site.

55. The content information analyzing apparatus as set forth in claim 54, wherein said means for obtaining comprises means for generating a plurality of log records, each said log records including a search keyword and a jump destination, from log data stored in a proxy server.

56. The content information analyzing apparatus as set forth in claim 54, wherein said counter comprises:

means for normalizing the search keyword; and

means for deleting a search log other than a last search log if a plurality of search logs relating the same keyword within a predetermined time are included.

57. A program embodied on a medium for causing a computer to perform a content information analysis, said program comprising the steps of:

extracting a disclosure unit of an individual opinion from collected content information;

specifying an object of said individual opinion; and

analyzing a disclosed content of said individual opinion and specifying an evaluation as to said object of said individual opinion.

58. A program embodied on a medium for causing a computer to perform a content information analysis, said program comprising the steps of:

extracting a disclosure unit of an individual opinion from collected content information;

specifying an object of said individual opinion; and

determining reliability of said disclosure unit of said individual opinion.

59. A program embodied on a medium for causing a computer to perform a content information analysis, said program comprising the steps of:

obtaining a plurality of search logs for the content information, each said search log including a search keyword and a jump destination URL;

counting a number of accesses and a number of kinds of search keywords for each said jump destination URL in accordance with a predetermined rule;

totaling said number of accesses and said number of kinds of search keywords up for the jump destination URLs included in lower layers of each site specified based on a structure of the URL; and

identifying a type of said site based on the totaled number of accesses and the totaled number of kinds of search keywords for said site.