Organizing a collection of objects

• US 7,747,625 B2
• Filed: 07/31/2003
• Issued: 06/29/2010
• Est. Priority Date: 07/31/2003
• Status: Active Grant

• Current Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Original Assignee: Hewlett-Packard Development Company, L.P. (HP Inc.)
• Inventors: Gargi, Ullas, Tretter, Daniel R.
• Primary Examiner(s): Pham, Hung Q
• Application Number:
  US10/631,369
• Publication Number:
  US 20050027712A1
• Time in Patent Office: 2,525 Days
• Field of Search: 707/100, 707/103.R, 707/999.002, 707/999.003, 707/999.004, 707/999.007, 707/999.01
• US Class Current: 707/737
• CPC Class Codes: G06F 16/54 Browsing; Visualisation therefor
• CPC Class Codes: G06F 16/58 Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually

Abstract

Systems and methods of organizing a collection of objects are described. In one aspect, a sequence of objects is segmented into object clusters based on: comparisons of successive object intervals to weighted measures of cluster extent; and comparisons of successive object intervals to weighted measures of cluster object density. In another aspect, objects from the collection are segmented into clusters. Context-related meta data associated with the objects and parsable into multiple levels of a name hierarchy is extracted. Names are assigned to clusters based on the extracted context-related meta data corresponding to a level of the name hierarchy selected to distinguish segmented clusters from one another. In another aspect, a sequence of objects that are segmented into clusters is accessed. Each cluster includes multiple objects arranged in a respective sequence in accordance with context-related meta data associated with the objects. At least two constituent objects representative of beginning and ending instances in the corresponding object sequence are selected for each object cluster. The selected representative objects of each cluster are graphically presented on a screen.

Claims

1. A method of organizing, a collection of objects arranged in a sequence ordered in accordance with a selected dimension of context-related metadata respectively associated with the objects, comprising operating a processor to perform operations comprising:

classifying the objects in the sequence to generate a series of object clusters, wherein the classifying comprises sequentially processing each of the objects as a respective candidate for segmentation into a respective current one of the object clusters in the series and, for each of the candidate objects,

determining a candidate object interval separating the candidate object from an adjacent object in the sequence already segmented into the current object cluster, the candidate object interval being measured in the selected dimension of the context-related metadata,

comparing the candidate object interval to a weighted measure of cluster extent for the current object cluster, the measure of cluster extent corresponding to a current distance spanned by all the objects in the current object cluster measured in the selected dimension of the context-related metadata, and

comparing the candidate object interval to a weighted measure of object density for the current object cluster, the measure of object density corresponding to a measure of distribution of distances separating adjacent ones of the objects in the current object cluster measured in the selected dimension of the context-related metadata.

2. The method of claim 1, wherein the measure of cluster extent for each current object cluster corresponds to a temporal distance spanned by recorded generation times associated with all objects in the current object cluster.

3. The method of claim 1, wherein the measure of cluster extent for each current object cluster corresponds to a spatial distance spanned by recorded generation locations associated with all objects in the current object cluster.

4. The method of claim 1, wherein the measure of object density for each current object cluster corresponds to an average temporal distance separating adjacent objects in the current object cluster.

5. The method of claim 1, wherein the measure of object density for each current object cluster corresponds to an average spatial distance separating adjacent objects in the current object cluster.

6. The method of claim 1, wherein the classifying comprises merging consecutive ones of the candidate objects into a current one of the object clusters until the candidate object interval determined for a current one of the candidate objects exceeds the weighted measure of cluster extent for the current cluster, at which point a successive one of the object clusters in the series is initiated with the current candidate object.

7. The method of claim 1, wherein the classifying comprises merging consecutive ones of the candidate objects into a current one of the object clusters until the candidate object interval determined for a current one of the candidate objects exceeds the weighted measure of object density for the current object cluster, at which point a successive one of the object clusters in the series is initiated with the current candidate object.

8. The method of claim 1, wherein the processing comprises determining the weighted measures of cluster extent by applying to the measures of cluster extent respective weights that decrease with increasing sizes of the respective object clusters.

9. The method of claim 1, wherein the processing comprises determining the weighted measures of cluster extent by applying to the measures of cluster extent respective weights that decrease with increasing sizes of the respective object clusters.

10. The method of claim 1, further comprising customizing at least one of the weights applied to the measures of cluster extent based on an analysis of objects in the corresponding object cluster.

11. The method of claim 10, wherein the customizing comprises scaling at least one of the weights applied to the measures of cluster extent based on a fractal dimension estimate of recorded time generation meta data associated with the objects in the collection.

12. The method of claim 1, further comprising customizing at least one of the weights applied to the measures of cluster object density based on an analysis of objects in the corresponding object cluster.

13. The method of claim 12, wherein the customizing comprises scaling at least one of the weights applied to the measures of cluster extent based on a fractal dimension estimate of recorded time generation meta data associated with the objects in the collection.

14. The method of claim 1, wherein the processing further comprises comparing the object density of a candidate object cluster consisting of the current object cluster and the candidate object with the weighted measure of object density for the current object cluster.

15. The method of claim 14, wherein the measure of object density for each current object cluster corresponds to an average temporal distance separating adjacent objects in the current object cluster.

16. The method of claim 14, wherein the measure of object density for each current object cluster corresponds to an average spatial distance separating adjacent objects in the current object cluster.

17. The method of claim 14, wherein the measure of object density for each object cluster corresponds to a moving average distance separating adjacent objects in the current object cluster.

18. The method of claim 14, wherein the processing comprises determining the weighted measures of cluster extent by applying to the measures of cluster extent respective weights that decrease with increasing sizes of the respective object clusters.

19. The method of claim 1, wherein the processing comprises processing each of the candidate objects sequentially beginning at a first end of the object sequence.

20. The method of claim 19, wherein the processing further comprises processing each of the candidate objects sequentially beginning at a second end of the object sequence opposite the first end.

21. The method of claim 1, wherein the sequence to be segmented includes objects of the following types:

text, audio, graphics, still images, video and business events.

22. A system of organizing a collection of objects arranged in a sequence ordered in accordance with a selected dimension of context-related metadata respectively associated with the objects, comprising:

a computer-readable medium storing computer-readable instructions; and

a data processing unit coupled to the computer-readable medium operable to execute the instructions, and based at least in part on the execution of the instructions operable to perform operations comprising

classifying the objects in the sequence to generate a series of object clusters, wherein each of the objects is sequentially processed as a respective candidate for segmentation into a respective current one of the object clusters in the series and, for each of the candidate objects, perform operations comprising

determining a candidate object interval separating the candidate object from an adjacent object in the sequence already segmented into the current object cluster, the candidate object interval being measured in the selected dimension of the context-related metadata,

compare the candidate object interval to a weighted measure of cluster extent for the current object cluster, the measure of cluster extent corresponding to a current distance spanned by all the objects in the current object cluster measured in the selected dimension of the context-related metadata, and

comparing the candidate object interval to a weighted measure of cluster object density for the current object cluster, the measure of object density corresponding to a measure of distribution of distances separating adjacent ones of the objects in the current object cluster measured in the selected dimension of the context-related metadata.

23. A method of organizing a collection of objects, comprising operating a processor to perform operations comprising:

segmenting objects from the collection into clusters;

extracting context-related meta data corresponding to object generation locations associated with the objects, wherein the extracted context-related meta data are parsable into multiple levels of a geographical name hierarchy generated for browsing; and

assigning location names to clusters, wherein each cluster is assigned a respective location name based on the associated extracted context-related meta data corresponding to a level of the geographical name hierarchy selected to distinguish segmented clusters from one another.

24. The method of claim 23, wherein the context-related meta data corresponds to recorded information relating to country, city, and state of object generation.

25. The method of claim 23, wherein the context-related meta data corresponds to both object generation times and object generation locations.

26. The method of claim 23, further comprising automatically naming objects in a given cluster based on the location name assigned to the given cluster.

27. The method of claim 26, wherein the objects in the given cluster are named automatically in accordance with a chronological ordering of the objects in the given cluster.

28. The method of claim 26, further comprising storing objects in the given cluster in a tree structure organized by cluster and labeled in accordance with the assigned names.

29. A system of organizing a collection of objects, comprising:

a computer-readable medium storing computer-readable instructions; and

a data processing unit coupled to the computer-readable medium, operable to execute the instructions, and based at least in part on the execution of the instructions operable to perform operations comprising

segmenting objects from the collection into clusters; and

extracting context-related meta data corresponding to object generation locations associated with the objects, wherein the extracted context-related meta data are parsable into multiple levels of a geographical name hierarch generated for browsing; and

assigning location names to each cluster, wherein each cluster is assigned a respective location name based on the associated extracted context-related meta data corresponding to a level of the geographical name hierarchy selected to distinguish segmented clusters from one another.

30. A method of organizing a collection of objects, comprising operating a processor to perform operations comprising:

accessing a sequence of objects segmented into object clusters each including multiple constituent objects arranged in a respective sequence in accordance with context-related meta data associated with the objects;

for each object cluster selecting at least two constituent objects representative of beginning and ending instances in the corresponding constituent object sequence in the object cluster, wherein the at least two constituent object comprising first and last constituent objects in the corresponding constituent object sequence of the object cluster; and

in a user interface, graphically presenting the selected constituent objects of each object cluster without graphically presenting representations of unselected ones of the constituent objects of the object clusters.

31. The method of claim 30, further comprising graphically presenting a selected one of the object clusters as a stack of partially overlapping images representative of multiple objects in the selected object cluster.

32. The method of claim 31, further comprising revealing an increased portion of a given one of the representative images in the stack in response to detection of a user-controlled display icon positioned over the given representative image.

33. The method of claim 30, wherein the presenting comprises presenting the selected constituent objects with the spacing between adjacent ones of the selected constituent objects in the same object cluster smaller than the spacing between adjacent ones of the selected constituent objects in different object clusters.

34. The method of claim 30, further comprising merging objects of one object cluster into an adjacent object cluster in response to user input.

35. The method of claim 34, wherein objects of one object cluster are merged into an adjacent object cluster in response to dragging and dropping of the objects to be merged.

36. The method of claim 34, wherein the objects of the one object cluster are merged into the adjacent object cluster in response to user selection of an icon for merging the object clusters.

37. The method of claim 30, further comprising presenting a graphical representation of distributions of objects in the object clusters.

38. The method of claim 37, wherein a object distribution for a given object cluster is presented as object instances plotted along an axis corresponding to a scaled representation of the context-related extent spanned by the given object cluster.

39. The method of claim 37, further comprising splitting a given object cluster in response to user selection of a point in the representation of the object distribution presented for the given object cluster.

40. The method of claim 37, further comprising automatically splitting a given object cluster into two or more object clusters in response to user input.

41. The method of claim 40, wherein the given object cluster is automatically split into a user-selected number of sub-clusters.

42. The method of claim 40, wherein the given object cluster is automatically split based on relative sizes of intervals between successive objects in the given object cluster.

43. The method of claim 30, wherein the context-related meta data corresponds to object generation times.

44. The method of claim 30, wherein the context-related meta data corresponds to object generation locations.

45. The method of claim 30, wherein the segmented sequence includes objects of the following types:

text, audio, graphics, still images, video, and business events.

46. The method of claim 30, further comprising graphically presenting at least one link to an object of an object cluster arranged in a sequence in accordance with time-related meta data in a calendar format.

47. The method of claim 30, further comprising graphically presenting at least one link to an object of an object cluster arranged in a sequence in accordance with location-related meta data in a map format.

48. A system of organizing a collection of objects, comprising:

a computer-readable medium storing computer-readable instructions; and

a data processing unit coupled to the computer-readable medium, operable to execute the instructions, and based at least in part on the execution of the instructions operable to perform operations comprising

accessing a sequence of objects from the collection segmented into clusters each including multiple objects arranged in a respective sequence in accordance with context-related meta data associated with the objects;

for each object cluster selecting at least two constituent objects representative of beginning and ending instances in the corresponding constituent object sequence in the object cluster, wherein the at least two constituent objects comprising first and last constituent objects in the corresponding constituent object sequence of the object cluster; and

in a user interface, graphically presenting the selected constituent objects of each object cluster on a screen without graphically presenting representations of unselected ones of the constituent objects of the object clusters, wherein the selected constituent objects are presented with the spacing between adjacent ones of the selected constituent objects in the same object cluster smaller than the spacing between adjacent ones of the selected constituent objects in different object clusters.