System and method for multi-dimensional foraging and retrieval of documents

• US 20050138056A1
• Filed: 12/18/2003
• Published: 06/23/2005
• Est. Priority Date: 12/18/2003
• Status: Active Grant

• Current Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Original Assignee: Xerox Corporation (Xerox Holdings Corp.)
• Inventors: Stefik, Mark J., Brawer, Alexander B.
• Granted Patent: US 7,707,210 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/3328 using graphical result space presentation or visualisation
• CPC Class Codes: G06F 16/93 Document management systems

Abstract

A method for interactive foraging for documents within a corpus of documents includes forming a working set of documents which are a subset of the corpus of documents. A visual representation of the working set and a portion of the corpus other than the working set are generated, and a visual representation of a retrieval boundary, between the working set and the portion of the corpus other than the working set, is presented. A further embodiment provides a system for interactive foraging of documents within a corpus of documents which includes, a visualization foraging subsystem, and an analytical repository containing a plurality of documents. The documents have been analyzed and are provided with associated dimensional data. A search mechanism of the visualization foraging subsystem is configured to form a working set of documents from the corpus of documents. A display device displays the working set and a portion of the corpus other than the working set, and a retrieval boundary is displayed on the display distinguishing between documents in the working set and the displayed documents of the corpus other than the working set.

Claims

1. A method for interactive foraging for documents within a corpus of documents comprising:

forming a working set of documents which are a subset of the corpus;

generating a visual representation of the working set and at least a portion of the corpus other than the working set; and

generating a visual representation of a retrieval boundary between the working set and the portion of the corpus other than the working set.

2. The method as set forth in claim 1, wherein the working set is a dynamically alterable working set, and altering the working set repositions the retrieval boundary.

3. The method as set forth in claim 1, wherein the portion of the corpus other than the working set, are neighbor documents having a relationship to the documents of the working set.

4. The method as set forth in claim 1, wherein the step of forming the working set includes supplying at least one search term to access documents of the corpus.

5. The method as set forth in claim 1, wherein the step of forming the working set includes manually selecting by a user at least one document of the corpus.

6. The method as set forth in claim 1, wherein the step of forming the working set includes using a previously generated working set.

7. The method as set forth in claim 3, further including altering a composition of the working set, wherein at least one neighbor document becomes part of the working set.

8. The method as set forth in claim 7, wherein the altering step includes creating at least one new neighbor document.

9. The method as set forth in claim 3, further including altering a composition of the working set, wherein at least one document of the working set is changed to at least one neighbor document.

10. The method as set forth in claim 1, wherein the step of forming the working set includes including a set of documents having a similar value for a dimension.

11. The method as set forth in claim 10, wherein the working set includes documents with common or similar values in each of several dimensional data.

12. The method as set forth in claim 11, further including,

separating documents of the working set into document sets in accordance with dimensions associated with the documents;

and

displaying multiple views of the documents, wherein the multiple views are displayed on panels.

13. The method as set forth in claim 10, wherein dimensions of the documents of the working set are computed automatically.

14. The method as set forth in claim 10, wherein the computing of dimensional data occurs prior to a foraging operation or while the foraging operation is occurring.

15. The method as set forth in claim 1, further including,

altering, tentatively, a composition of the working set of documents by either adding or eliminating documents having a particular value in at least one dimension,

wherein the altering step alters at least one of the multiple views of the documents, within the panels.

16. The method as set forth in claim 1, wherein the step of forming the working set is accomplished by applying seeding operations, inclusion constraints, or exclusion constraints to a corpus.

17. A system for interactive foraging of documents within a corpus of documents, comprising:

a foraging subsystem;

an analytical repository containing a plurality of documents which have been analyzed and which have associated dimensional data;

a search mechanism of the foraging subsystem, the search mechanism configured to form a working set of documents from the corpus;

a display which displays the working set and at least a portion of the corpus other than the working set; and

a retrieval boundary displayed on the display to distinguish between documents in the working set and displayed documents of the corpus other than the working set.

18. The system as set forth in claim 17, wherein the working set is a dynamically alterable working set, and altering the working set repositions the retrieval boundary.

19. The system as set forth in claim 17, wherein the portion of the corpus other than the working set, include neighbor documents having a relationship to the documents of the working set.

20. A multi-dimensional foraging method comprising:

generating a working set of documents from a collection of documents having dimensions, wherein a dimension of a document represents a characteristic of the document;

selecting panels, each panel corresponding to a selected dimension or combination of dimensions of the documents;

generating for each of the selected panels, sets of the documents in the working set, the generated document sets corresponding to the dimension or combination of dimensions of a corresponding panel;

generating neighboring document sets from documents other than those of the working set, the neighboring document sets related to the document sets generated from the working set;

arranging the document sets having a particular dimension or combination of dimensions in relationship to each other in a same panel; and

displaying the panels having the arranged document set on a visual display device, wherein a multi-dimensional view of the working set is provided.

21. The method as set forth in claim 20 wherein the displaying step includes generating a visual retrieval boundary between document sets within the working set and document sets outside of the working set.

22. The method as set forth in claim 21 further including saving the retrieval boundary.

23. The method as set forth in claim 22 further including running the saved retrieval boundary over an altered collection of documents.

24. The method as set forth in claim 21 wherein the document sets are represented by visual tokens.

25. The method as set forth in claim 21 wherein the retrieval boundary is generated incrementally.

26. The method as set forth in claim 20 wherein adding at least one of a constraint or inclusion to at least one of the displayed panels causes an altering of the working set.

27. The method as set forth in claim 20 further comprising altering the view of the multi-dimensional view of the working set by adding at least one of a constraint or inclusion to at least one of the displayed panels.

28. The method as set forth in claim 27 wherein the adding of at least one of the constraint or inclusion in a first panel alters a view generated in another panel.

29. The method as set forth in claim 20 wherein the step of generating neighboring document sets includes performing at least one of a nearness operation or a degree of interest operation, from document sets generated from the working set.

30. In a data communication network a multi-dimensional retrieval system comprising:

a working set of documents generated from a collection of documents having dimensions, wherein a dimension of a document represents a characteristic of the document;

a plurality of panels, each panel corresponding to a selected dimension or combination of dimensions of the documents;

a plurality of documents generated from the documents in the working set, the generated documents representative of the dimension or combination of dimensions for a corresponding panel;

a plurality of neighboring document sets formed from documents other than the documents of the working set, the neighboring documents related to the documents generated from the working set; and

a display for displaying multiple dimensions, including at least some of the plurality of panels having ones of the document sets, the document sets corresponding to the dimension or combination of dimensions of the respective panel, and the selected document sets arranged in a relationship to each other.

31. The data communication network according to claim 30 further including a client/server configuration wherein communication between the client and server employs a communications protocol that supports computations based on incremental communication of search results.

32. The data communication network according to claim 31 wherein the protocol includes a prioritizing communications protocol which prioritizes communications to first communicate results within the working set.

33. The data communication network according to claim 31 wherein the protocol includes a summarizing communications protocol which summarizes or provides approximations of results prior to communicating details of the results.

34. The data communication network according to claim 30 further including an analysis process configured to run concurrent with and independent of operation of the system by a user, and wherein findings of the analysis process are cached in a back-end repository.