Information storage and retrieval

• US 7,668,853 B2
• Filed: 11/25/2003
• Issued: 02/23/2010
• Est. Priority Date: 11/27/2002
• Status: Expired due to Fees

• Current Assignee: Sony United Kingdom Limited (Sony Group Corp.)
• Original Assignee: Sony United Kingdom Limited (Sony Group Corp.)
• Inventors: Thorpe, Jonathan Richard, Trepess, David William
• Primary Examiner(s): Fleurantin; Jean B.
• Assistant Examiner(s): COLAN, GIOVANNA B
• Application Number:
  US10/723,422
• Publication Number:
  US 20040139105A1
• Time in Patent Office: 2,282 Days
• Field of Search: 707/104.1, 707/10, 707/102, 707/104, 707/3, 707/7, 707/5, 707/2, 707/1, 725/116, 704/9
• US Class Current: 707/717
• CPC Class Codes: G06F 16/338 Presentation of query results
• CPC Class Codes: G06F 16/355 Class or cluster creation or modification
• CPC Class Codes: G06F 16/358 Browsing; Visualisation therefor

Abstract

An information retrieval system in which a set of distinct information items map to respective nodes in an array of nodes by mutual similarity of the information items, so that similar information items map to nodes at similar positions in the array of nodes; the system comprising: a user control for defining a search criterion for selecting information items; a detector for detecting those positions within the array of nodes corresponding to the selected information items; a graphical user interface for displaying a two-dimensional display array of display points representing those positions within the array of nodes corresponding to the selected information items; and a processor responsive to the selected information items, for providing a representation which is generally representative of the information content of the selected information items.

Claims

1. An information retrieval system in which a set of distinct information items map to respective nodes in a self-organizing map by mutual similarity of the information items, so that similar information items map to nodes at similar positions in the self-organizing map, wherein the self-organizing map is trained upon reduced dimension characterizations of the information items, the system comprising:

a user control configured to define a search criterion for selecting a subset of the information items represented by the self-organizing map, the search criterion being applied using a standard keyword search technique;

a detector configured to detect those positions within the self-organizing map corresponding to the subset of the information items selected by the standard keyword search technique;

a graphical user interface configured to display points representing only those positions within the self-organizing map corresponding to the selected subset of information items; and

a processor, responsive to the selected subset of information items selected using the search criterion, configured to provide one or more representations representative of the information content of the selected subset of information items,

wherein each of the information items represented by the self-organizing map includes image data; and

wherein the processor is responsive to the selected subset of information items and causes the graphical user interface to display one or more images obtained from the image data included in the selected subset of information items selected using the search criterion so as to represent the content of the selected subset of information items.

2. A system according to claim 1, wherein the graphical user interface is configured to display a two-dimensional display array of the said display points.

3. A system according to claim 2, in which a dither component is applied to the mapping between information items and nodes in the self-organizing map so that information items that share a node tend to map to closely spaced, but different positions in the displayed array.

4. A system according to claim 2, in which the information items are mapped to nodes in the self-organizing map on the basis of a feature vector derived from each information item.

5. A system according to claim 4, in which the feature vector for an information item represents a set of frequencies of occurrence, within that information item, of each of a group of information features.

6. A system according to claim 5, in which the information items comprise textual information, the feature vector for an information item represents a set of frequencies of occurrence, within that information item, of each of a group of words.

7. A system according to claim 1, in which the information items comprise textual information, the nodes being mapped by mutual similarity of at least a part of the textual information.

8. A system according to claim 6, in which the information items are pre-processed for mapping by excluding words occurring with more than a threshold frequency amongst the set of information items.

9. A system according to claim 6, in which the information items are pre-processed for mapping by excluding words occurring with less than a threshold frequency amongst the set of information items.

10. A system according to claim 1, wherein the said user control comprises:

search means for carrying out a search of the information items represented by the self-organizing map;

the search means and the graphical user interface being configured to co-operate so that only those display points corresponding to the subset of information items selected by the search means are displayed on a user display.

11. A system according to claim 1, wherein the said processor is configured to detect clusters of similar information items and to provide representations representative of the information content of the respective clusters.

12. A system according to claim 1, wherein the processor is configured to provide each representation on a user display as a label of the display points corresponding to the information items represented thereby.

13. A system according to claim 12, wherein the label is a word or set of words.

14. A system according to claim 11, wherein the processor determines, in respect of a set of information items with which a label is to be associated, the most frequently used word or set of words in the information items corresponding to the selected subset of the information items and applies that word or that set of words as the label.

15. A system according to claim 1, wherein the said processor is configured to select, from the image data, an image which is representative of the image data as a whole according to a predetermined selection criterion.

16. A system according to claim 1, wherein the processor is configured to select an image, a property of which is nearest to the average of a same property of the image data.

17. A system according to claim 1, wherein the said one or more images are applied as labels to the display points corresponding to the information items represented thereby.

18. A portable data processing device comprising a system according to claim 1.

19. Video acquisition and/or processing apparatus comprising a system according to claim 1.

20. An information retrieval method in which a set of distinct information items map to respective nodes in a self-organizing map by mutual similarity of the information items, so that similar information items map to nodes at similar positions in the self-organizing map, wherein the self-organizing map is trained upon reduced dimension characterizations of the information items, the method comprising:

defining a search criterion for selecting a subset of the information items represented by the self-organizing map, the search criterion being applied using a standard keyword search technique;

detecting those positions within the self-organizing map corresponding to the subset of the information items selected by the standard keyword search technique;

displaying display points which are at positions representing only those positions within the self-organizing map corresponding to the selected subset of information items; and

in response to the selected subset of information items selected using the search criterion, providing one or more representations representative of the information content of the selected subset of information items,

wherein each of the information items represented by the self-organizing map includes image data; and

wherein the providing step includes displaying one or more images obtained from the image data included in the selected subset of information items selected using the search criterion so as to represent the content of the selected subset of information items.

21. A method according to claim 20, wherein the displaying step displays a two-dimensional display array of the said display points.

22. A method according to claim 20, comprising:

carrying out a search of the information items;

displaying on a user display only those display points corresponding to the subset of information items selected by the search.

23. A method according to claim 20, further comprising detecting clusters of similar information items and providing representations representative of the information content of the respective clusters.

24. A method according to claim 20, further comprising providing each representation on a user display as a label of the display points corresponding to the information items represented thereby.

25. A method according to claim 24, wherein the label is a word or set of words.

26. A method according to claim 20, further comprising providing the one or more images representative of the information content of the selected subset of information items selected using the search criterion.

27. A method according to claim 26, further comprising selecting, from the image data, an image which is representative of the image data as a whole according to a predetermined selection criterion.

28. A method according to claim 27, further comprising selecting the image, a property of which is nearest to the average of the same property of said image data.

29. A computer-readable medium storing a program which, when executed by a computer, causes the computer to perform the method recited in claim 20.

30. A computer-readable medium storing a program that, when executed by a computer, causes the computer to generate a user interface of an information retrieval system in which a set of distinct information items map to respective nodes a self-organizing map by mutual similarity of the information items, so that similar information items map to nodes at similar positions in the self-organizing map, wherein the self-organizing map is trained upon reduced dimension characterizations of the information items, the interface comprising:

a user control for defining a search criterion for selecting a subset of the information items represented by the self-organizing map, the search criterion being applied using a standard keyword search technique; and

a graphical user interface configured to display points representing only those positions within the self-organizing map corresponding to the subset of information items selected by the standard keyword search technique and to display one or more representations representative of the information content of the subset of information items selected by the search criterion,

wherein each of the information items represented by the self-organizing maps includes image data; and

wherein the graphical user interface is configured to display one or more images obtained from the image data included in the selected subset of information items selected using the search criterion so as to represent the content of the selected subset of information items.

31. A user interface according to claim 30, wherein the said user control comprises:

search means for carrying out a search of the information items represented by the self-organizing map;

the search means and the graphical user interface being configured to co-operate so that only those display points corresponding to the subset of information items selected by the search are displayed on the user display.

32. An interface according to claim 30, wherein the graphical user interface is configured to display representations representative of the information content of respective clusters of similar information items.

33. An interface according to claim 30, wherein graphical user interface is configured to provide each representation as a label of the display points corresponding to the information items represented thereby.

34. An interface according to claim 33, wherein the label is a word or set of words.

35. An interface according to claim 30, wherein the said representations are images that are applied as labels to the display points corresponding to the information items represented thereby.