Method and system for knowledge repository exploration and visualization

• US 20030004932A1
• Filed: 06/20/2001
• Published: 01/02/2003
• Est. Priority Date: 06/20/2001
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Chow, Amy W., Proctor, Larry L., Spangler, W. Scott, Kreulen, Jeffrey T., Lessler, Justin T.
• Granted Patent: US 6,725,217 B2
• Time in Patent Office: Days
• US Class Current: 707/3
• CPC Class Codes: G06F 16/3328 using graphical result space presentation or visualisation
• CPC Class Codes: G06F 16/954 Navigation, e.g. using categorised browsing
• CPC Class Codes: Y10S 707/99936 Pattern matching access
• CPC Class Codes: Y10S 707/99943 Generating database or data structure, e.g. via user interface

Abstract

A computing system and method explores a knowledge repository by accepting a natural language query from a user, determining a distance between the query and every category in every partitioning of the knowledge repository, and displaying a radial graph (322) of the nearest categories. Further, in response to a user selecting a category, visually displaying matching elements in the category along with its nearest neighbor categories in a scatter plot (324).

Claims

1. A method comprising the steps of:

receiving a natural language query;

calculating the distance between the query and a plurality of categories in partitions of a knowledge repository; and

presenting at least one category that comprises a nearest distance to the query.

2. The method of claim 1, further comprising a step of:

displaying the at least one category via radial graph.

3. The method of claim 2, wherein the radial graph comprises:

a keyword hub;

a set of weighted spokes, radially coupled to the keyword hub; and

a set of related concept boxes, each related concept box being coupled to an end of a weighted spoke.

4. The method of claim 3, wherein each of the set of weighted spokes comprises a relevance factor indicated by its style and color.

5. The method of claim 1, further comprising the steps of:

selecting a category; and

displaying matching knowledge elements in a selected category along with nearest neighbor categories.

6. The method of claim 5, wherein the displaying step comprises a scatter plot representation.

7. The method of claim 5, further comprising the step of displaying terms from a dictionary associated with a selected category.

8. The method of claim 5, wherein the knowledge element is an electronic document.

9. The method of claim 5, further comprising the steps of:

detecting a selection of one point in the scatter plot representation;

searching text in a knowledge element associated with the selected one point until a first occurrence of a user query is found;

displaying a query term at least in partial character form;

selecting a knowledge element; and

presenting the selected knowledge element.

10. The method of claim 9, further comprising the step of highlighting all occurrences of query text within the selected knowledge element.

11. A system comprising:

a visual user interface;

a visual user interface/event manager communicatively coupled to the visual user interface;

a query handler, communicatively coupled to the visual user interface/event manager;

an application programming interface, communicatively coupled to the visual user interface/event manager;

a look-up system, communicatively coupled to the application programming interface;

a vector calculator, communicatively coupled to the application programming interface;

a result set, communicatively coupled to the vector calculator; and

a result set manager, communicatively coupled to the visual user interface/event manager, to the application programming interface, and to the result set, and wherein the visual user interface/event manager, in response to receiving a natural language query, calculates the distance between the query and a plurality of categories in partitions of a knowledge repository, and presents, via a radial graph in the visual user interface, at least one category that comprises a nearest distance to the query.

12. The system of claim 11, wherein the radial graph comprises:

a keyword hub;

a set of weighted spokes, radially coupled to the keyword hub; and

a set of related concept boxes, each related concept box being coupled to an end of a weighted spoke.

13. The system of claim 12, wherein each of the set of weighted spokes comprises a relevance factor indicated by its style and color.

14. The system of claim 11, wherein the look-up system comprises

a knowledge database;

a classification record, communicatively coupled to the knowledge database;

a dictionary space, communicatively coupled to the classification record and to the knowledge database;

a vector space, communicatively coupled to the knowledge database, the classification record, and the dictionary space; and

a centroid record, communicatively coupled to the knowledge database, the classification record, the dictionary space, and the vector space.

15. The system of claim 14, wherein the knowledge database comprises multiple partitionings based on different aspects of their content.

17. The computer readable medium of claim 16, further comprising instructions for a step of:

displaying the at least one category via radial graph.

18. The computer readable medium of claim 17, wherein the radial graph comprises:

a keyword hub;

a set of weighted spokes, radially coupled to the keyword hub; and

a set of related concept boxes, each related concept box being coupled to an end of a weighted spoke.

19. The computer readable medium of claim 18, wherein each weighted spoke comprises a relevance factor indicated by its style and color.

20. The computer readable medium of claim 16, further comprising computer instructions for:

selecting a category; and

displaying matching knowledge elements in a selected category along with nearest neighbor categories in a scatter plot representation.

21. The computer readable medium of claim 20, wherein the knowledge elements are electronic documents.

22. The computer readable medium of claim 16, further comprising computer instructions for:

detecting a selection of one point in the scatter plot representation;

searching text in a knowledge element associated with the selected one point until a first occurrence of a user query is found;

displaying a query term at least in partial character form;

selecting a knowledge element; and

presenting the selected knowledge element.

24. The apparatus of claim 23, wherein the look-up system comprises:

a knowledge database;

a classification record, communicatively coupled to the knowledge database;

a dictionary space, communicatively coupled to the classification record and to the knowledge database;

a vector space, communicatively coupled to the knowledge database, the classification record, and the dictionary space; and

a centroid record, communicatively coupled to the knowledge database, the classification record, the dictionary space, and the vector space.

16. A computer readable medium including computer instructions for driving a visual user interface, the computer instructions comprising instructions for:

receiving a natural language query;

calculating the distance between the query and a plurality of categories in partitions of a knowledge repository; and

presenting at least one category that comprises a nearest distance to the query.

23. An apparatus comprising:

a visual user interface;

a visual user interface/event manager communicatively coupled to the visual user interface;

a query handler, communicatively coupled to the visual user interface/event manager;

an application programming interface, communicatively coupled to the visual user interface/event manager;

a look-up system, communicatively coupled to the application programming interface;

a vector calculator, communicatively coupled to the application programming interface;

a result set, communicatively coupled to the vector calculator; and

a result set manager, communicatively coupled to the visual user interface/event manager, to the application programming interface, and to the result set for;

receiving a natural language query;

calculating the distance between the query and a plurality of categories in partitions of a knowledge repository; and

displaying, via a radial graph in the visual user interface, at least one category that comprises a nearest distance to the query.