Method of, apparatus for, and computer program for mapping contents having meta-information

• US 20030093432A1
• Filed: 11/15/2002
• Published: 05/15/2003
• Est. Priority Date: 11/16/2001
• Status: Active Grant

• Current Assignee: Nippon Telegraph and Telephone Corporation
• Original Assignee: Nippon Telegraph and Telephone Corporation
• Inventors: Abe, Shinji, Hayashi, Yasuhito, Miyahara, Shinji, Fujita, Etsuro
• Granted Patent: US 7,363,311 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/287 Visualization; Browsing
• CPC Class Codes: Y10S 707/99942 Manipulating data structure, e.g. compression, compaction, compilation
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

A method of mapping a large number of items of contents each having meta-information to a low dimensional space so that the map of the contents reflects the meta-information. The method includes the steps of assigning a concept vector to each item of contents, adjusting the distance between two items of contents so that the closer the classification information of the two items of contents match, the more the distance is reduced, and assigning position information in the low dimensional space to each item of contents based on the adjusted distance.

Claims

1. A method of mapping contents to a low dimensional space, each of a plurality of items of contents having meta-information including classification information, comprising the steps of:

assigning each item of contents a concept vector; and

assigning each item of contents position information in a low dimensional space based on the concept vector assigned to the item of contents;

wherein

two items of contents are mapped at a short distance from each other to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents.

2. The method as claimed in claim 1,

wherein the step of assigning the position information in the low dimensional space further comprises the steps of:

calculating a distance between two items of contents based on the concept vectors assigned thereto;

adjusting said distance based on the classification information of said two items of contents so that said distance is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents; and

assigning each item of contents the position information in the low dimensional space based on the adjusted distance.

3. The method as claimed in claim 2, wherein

the classification information is a classification hierarchy in which each item of contents is assigned a classification category;

and

said distance is reduced to such an extent that a number of levels at which the classification categories assigned to said two items of contents match.

4. The method as claimed in claim 3, wherein said distance between said two items of contents is adjusted by adding an adjustment value corresponding to the level down to which the classification categories of said two items of contents match.

5. The method as claimed in claim 3, wherein said distance between said two items of contents is adjusted by multiplying an adjustment coefficient corresponding to the level down to which the classification categories of said two items of contents match.

6. The method as claimed in claim 5, wherein said adjustment coefficient is calculated using a distance matrix of which size is determined by the number of the classification categories in the lowest level down to which classification categories of said two items of contents match.

7. The method as claimed in claim 1,

wherein the step of assigning the position information in the low dimensional space further comprises the steps of:

calculating a distance between two items of contents based on the concept vectors assigned thereto;

assigning each item of contents said position information in a low dimensional space based on the distance; and

adjusting said position information based on the classification information of said two items of contents so that the distance between the position information of said two items is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents.

8. The method as claimed in claim 7, wherein

classification information is a classification hierarchy in which each item of contents is assigned a classification category;

and

said distance is adjusted so that the closer the classification information of said two items of contents match, the shorter the distance between the position information of said two items of contents becomes.

9. The method as claimed in claim 8, wherein a barycenter of contents classified into a classification category of a level is calculated in response to a user request;

and

a distance between each item of contents and said barycenter is proportionally reduced.

10. The method as claimed in claim 1,

wherein the step of assigning the position information in the low dimensional space further comprises the steps of:

calculating a distance between two items of contents based on the concept vectors assigned thereto;

adjusting said distance based on the classification information of said two items of contents so that said distance is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents;

assigning each item of contents the position information in a low dimensional space based on the adjusted distance; and

adjusting said position information based on the classification information of said two items of contents so that the distance between the position information of said two items of the contents is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents.

11. The method as claimed in claim 2, wherein said position information in a low dimensional space is assigned to each item of contents by applying a multi-dimensional scaling method.

12. The method as claimed in claim 7, wherein said position information in a low dimensional space is assigned to each item of contents by applying a multi-dimensional scaling method.

13. The method as claimed in claim 10, wherein said position information in a low dimensional space is assigned to each item of contents by applying a multi-dimensional scaling method.

14. The method as claimed in claim 2, further comprising the step of transmitting said position information assigned to each item of contents and a classification category of the item of contents to an exterior via a network.

15. The method as claimed in claim 7, further comprising the step of transmitting said position information assigned to each item of contents and a classification category of the item of contents to an exterior via a network.

16. The method as claimed in claim 10, further comprising the step of transmitting said position information assigned to each item of contents and a classification category of the item of contents to an exterior via a network.

17. The method as claimed in claim 14, wherein a range of items of contents of which said position information and said classification categories are transmitted is expanded or reduced in a continuous manner in response to a user request.

18. The method as claimed in claim 15, wherein a range of items of contents of which said position information and said classification categories are transmitted is expanded or reduced in a continuous manner in response to a user request.

19. The method as claimed in claim 16, wherein a range of items of contents of which said position information and said classification categories are transmitted is expanded or reduced in a continuous manner in response to a user request.

20. An apparatus for mapping contents to a low dimensional space, each of a plurality of items of contents having meta-information including classification information, comprising:

means for assigning each item of contents a concept vector; and

means for assigning each item of contents position information in a low dimensional space based on the concept vector assigned to the item of contents;

wherein

two items of contents are mapped at a short distance from each other to such an extent that classification information of said one item of contents is close to the classification information of said other item of contents.

21. The apparatus as claimed in claim 20,

wherein said means for assigning the position information in the low dimensional space further comprises:

means for calculating a distance between two items of contents based on the concept vectors assigned thereto;

means for adjusting said distance based on the classification information of said two items of contents so that said distance is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents; and

means for assigning each item of contents the position information in the low dimensional space based on the adjusted distance.

22. The apparatus as claimed in claim 21, wherein

the classification information is a classification hierarchy in which each item of contents is assigned a classification category;

and

said distance is reduced to such an extent that a number of levels at which the classification categories assigned to said two items of contents match.

23. The apparatus as claimed in claim 22, wherein said distance between said two items of contents is adjusted by adding an adjustment value corresponding to the level down to which the classification categories of said two items of contents match.

24. The apparatus as claimed in claim 22, wherein said distance between said two items of contents is adjusted by multiplying an adjustment coefficient corresponding to the level down to which the classification categories of said two items of contents match.

25. The apparatus as claimed in claim 24, wherein said adjustment coefficient is calculated using a distance matrix of which size is determined by the number of the classification categories in the lowest level down to which classification categories of said two items of contents match.

26. The apparatus as claimed in claim 20,

wherein said means for assigning position information in a low dimensional space further comprises:

means for calculating a distance between two items of contents based on the concept vectors assigned thereto;

means for assigning each item of contents said position information in a low dimensional space based on the distance; and

means for adjusting said position information based on the classification information of said two items of contents so that distance between the position information of said two items is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents.

27. The apparatus as claimed in claim 26, wherein

the classification information is a classification hierarchy in which each item of contents is assigned a classification category;

and

said distance is adjusted so that the closer the classification information of said two items of contents match, the shorter the distance between the position information of said two items of contents becomes.

28. The apparatus as claimed in claim 27, wherein a barycenter of contents classified into a classification category of a level is calculated in response to a user request;

and

a distance between each item of contents and said barycenter is proportionally reduced.

29. The apparatus as claimed in claim 20,

wherein said means for assigning the position information in the low dimensional space further comprises:

means for calculating a distance between two items of contents based on the concept vectors assigned thereto;

means for adjusting said distance based on the classification information of said two items of contents so that said distance is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents;

means for assigning each item of contents the position information in a low dimensional space based on the adjusted distance; and

means for adjusting said position information based on the classification information of said two items of contents so that the distance between the position information of said two items of contents is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents.

30. The apparatus as claimed in claim 21, wherein said position information in a low dimensional space is assigned to each item of contents by applying a multi-dimensional scaling method.

31. The apparatus as claimed in claim 26, wherein said position information in a low dimensional space is assigned to each item of contents by applying a multi-dimensional scaling method.

32. The apparatus as claimed in claim 29, wherein said position information in a low dimensional space is assigned to each item of contents by applying a multi-dimensional scaling method.

33. The apparatus as claimed in claim 21, further comprising means for transmitting said position information assigned to each item of contents and a classification category of the item of contents to an exterior via a network.

34. The apparatus as claimed in claim 26, further comprising means for transmitting said position information assigned to each item of contents and a classification category of the item of contents to an exterior via a network.

35. The apparatus as claimed in claim 29, further comprising means for transmitting said position information assigned to each item of contents and a classification category of the item of contents to an exterior via a network.

36. The apparatus as claimed in claim 33, wherein a range of items of contents of which said position information and said classification categories are transmitted is expanded or reduced in a continuous manner in response to a user request.

37. The apparatus as claimed in claim 34, wherein a range of items of contents of which said position information and said classification categories are transmitted is expanded or reduced in a continuous manner in response to a user request.

38. The apparatus as claimed in claim 35, wherein a range of items of contents of which said position information and said classification categories are transmitted is expanded or reduced in a continuous manner in response to a user request.

39. A computer program that causes a computer to map contents to a low dimensional space, each of a plurality of items of contents having meta-information including classification information, comprising the steps of:

assigning each item of contents a concept vector; and

assigning each item of contents position information in a low dimensional space based on the concept vector assigned to the item of contents;

wherein

two items of contents are mapped at a short distance from each other to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents.

40. The computer program as claimed in claim 39,

wherein the step of assigning the position information in the low dimensional space further comprises the steps of:

calculating a distance between two items of contents based on the concept vectors assigned thereto;

adjusting said distance based on the classification information of said two items of contents so that said distance is reduced to such an extent that classification information of said one item of contents is close to the classification information of said other item of contents; and

assigning each item of contents the position information in the low dimensional space based on the adjusted distance.

41. The computer program as claimed in claim 39,

wherein the step of assigning the position information in the low dimensional space further comprises the steps of:

calculating a distance between two items of contents based on the concept vectors assigned thereto;

assigning each item of contents said position information in the low dimensional space based on the distance; and

adjusting said position information based on the classification information of said two items of contents so that the distance between the position information of said two items is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents.

42. The computer program as claimed in claim 39,

wherein the step of assigning the position information in the low dimensional space further comprises the steps of:

calculating a distance between two items of contents based on the concept vectors assigned thereto;

adjusting said distance based on the classification information of said two items of contents so that said distance is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents;

assigning each item of contents the position information in a low dimensional space based on the adjusted distance; and

adjusting said position information based on the classification information of said two items of contents so that the distance between the position information of said two items of contents is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents.

43. A computer readable recording medium storing thereto a computer program that causes a computer to map contents to a low dimensional space, each of a plurality of items of contents having meta-information including classification information, comprising the steps of:

assigning each item of contents a concept vector; and

assigning each item of contents position information in a low dimensional space based on the concept vector assigned to the item of contents;

wherein

two items of contents are mapped at a short distance from each other to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents.

44. The computer readable recording medium as claimed in claim 43,

wherein the step of assigning the position information in the low dimensional space further comprises the steps of:

calculating a distance between two items of contents based on the concept vectors assigned thereto;

adjusting said distance based on the classification information of said two items of contents so that said distance is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents; and

assigning each item of contents the position information in the low dimensional space based on the adjusted distance.

45. The computer readable recording medium as claimed in claim 43,

wherein the step of assigning the position information in the low dimensional space further comprises the steps of:

calculating a distance between two items of contents based on the concept vectors assigned thereto;

assigning each item of contents said position information in the low dimensional space based on the distance; and

adjusting said position information based on the classification information of said two items of contents so that the distance between the position information of said two items is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents.

46. The computer readable recording medium as claimed in claim 43,

wherein the step of assigning the position information in the low dimensional space further comprises the steps of:

calculating a distance between two items of contents based on the concept vectors assigned thereto;

adjusting said distance based on the classification information of said two items of contents so that said distance is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents;

assigning each item of contents the position information in a low dimensional space based on the adjusted distance; and

adjusting said position information based on the classification information of said two items of contents so that the distance between the position information of said two items of contents is reduced to such an extent that the classification information of said one item of contents is close to the classification information of said other item of contents.