Systems and methods for building and implementing ontology-based information resources

• US 7,739,218 B2
• Filed: 08/16/2005
• Issued: 06/15/2010
• Est. Priority Date: 08/16/2005
• Status: Active Grant

• Current Assignee: International Business Machines Corporation
• Original Assignee: International Business Machines Corporation
• Inventors: Sow, Daby M., Nnebe, Ijeoma M., Fuller, Nicholas C. M., Arguello, Juan Fernando, Drissi, Youssef
• Primary Examiner(s): Sparks; Donald
• Assistant Examiner(s): Bharadwaj; Kalpana
• Application Number:
  US11/204,726
• Publication Number:
  US 20070043742A1
• Time in Patent Office: 1,764 Days
• Field of Search: 707/2, 707/10, 704/503, 704/207, 704/208, 708/313, 706/20, 706/12, 706/25, 706/61
• US Class Current: 706/61
• CPC Class Codes: G06F 16/367 Ontology

Abstract

Systems and methods are provided for building and implementing ontology-based information resources. More specifically, multi-user collaborative, semi-automatic systems and methods are provided for constructing ontology-based information resources that are shared by a community of users, wherein ontology categories evolve over time based on categorization rules that are specified by the community of users as well as categorization rules that are automatically learned from knowledge obtained as a result of multi-user interactions and categorization decisions.

Claims

1. A computer readable medium embodying instructions executable by a processor to perform a method for managing electronic information, the method comprising:

storing electronic files embodying the electronic information in a shared information repository that is shared by a community of client access devices, wherein the electronic files are categorized according to an ontology, wherein the ontology comprises an ontological model that is incrementally built and adapted over time based on multi-user interactions with the electronic information of the system; and

incrementally adapting the ontology over time based on categorization rules that are automatically learned from knowledge obtained as a result of the multi-user interactions, wherein the multi-user interactions include one of accepting an automatic categorization of the electronic information and specifying a category of the electronic information.

2. The computer readable medium of claim 1, the method further comprising:

maintaining a repository of categorization rules that are used for automatically categorizing an electronic file to be stored in the shared information repository;

applying machine learning methods to generate new categorization rules or update existing categorization rules in the repository of categorization rules.

3. The computer readable medium of claim 1, the method further comprising;

receiving a request via a network from a client access device to store an uncategorized electronic file in the shared information repository; and

prompting the client access device with an option of storing the electronic file according to one or more ontological categories or according to one or more user-specified ontological categories.

4. The computer readable medium of claim 3, the method further comprising automatically monitoring user interactions with an external information sources.

5. The computer readable medium of claim 4, the method further comprising automatically prompting the client access device with an option of storing and categorizing an electronic file that is accessed from the external information source.

6. The computer readable medium of claim 4, the method further comprising storing search patterns in a repository of search histories.

7. The computer readable medium of claim 6, the method further comprising:

determining preferred subjects of interest information for one or more the client access devices based on information stored in the repository of search histories; and

using the preferred subjects of interest information to identify newly published electronic files in one or more external information sources.

8. The computer readable medium of claim 7, the method further comprising automatically notifying the client access device of a newly published electronic file that correspond to the preferred subjects of interest.

9. The computer readable medium of claim 7, the method further comprising automatically storing the newly published electronic file in the shared information repository.

10. The computer readable medium of claim 1, wherein storing electronic files in a shared information repository further comprises allowing a client access device to associate unique metadata to one or more electronic files stored in the shared information repository.

11. The computer readable medium of claim 1, wherein storing electronic files in a shared information repository further comprises independently categorizing one or more sections of an electronic file for storage in the shared information repository.

12. The computer readable medium of claim 1, further comprising a middleware module connected to the shared information repository via a network for automatically subscribing to at least one external data sources to track currently published electronic information.

13. The computer readable medium of claim 10, wherein the metadata comprises a user'"'"'s unique preferred interest (UPI), wherein the UPI is not shared with other users and facilitates user searches based on the user'"'"'s unique interests.

14. A computer readable medium embodying instructions executable by a processor to perform a method for managing electronic information, the method comprising:

storing electronic files embodying the electronic information in a shared information repository that is shared by a community of client access devices, wherein the electronic files are categorized according to an ontology, wherein the ontology comprises an ontological model that is incrementally built and adapted to newly learned terminology over time based on multi-user interactions with the electronic information of the system; and

incrementally adapting the ontology over time based on categorization rules that are automatically learned from historical logs of the multi-user interactions, wherein the multi-user interactions include one of accepting an automatic categorization of the electronic information, derived from monitoring at least one of user search patterns and user categorization patterns, and specifying a category of the electronic information.

15. The computer readable medium of claim 14, the method further comprising:

maintaining a repository of categorization rules that are used for automatically categorizing an electronic file to be stored in the shared information repository; and

applying machine learning methods to generate new categorization rules or update existing categorization rules in the repository of categorization rules.

16. The computer readable medium of claim 14, the method further comprising:

receiving a request via a network from a client access device to store an uncategorized electronic file in the shared information repository; and

prompting the client access device with an option of storing the electronic file according to one or more ontological categories or according to one or more user-specified ontological categories.

17. The computer readable medium of claim 16, the method further comprising automatically monitoring user interactions with an external information sources.

18. The computer readable medium of claim 17, the method further comprising automatically prompting the client access device with an option of storing and categorizing an electronic file that is accessed from the external information source.

19. The computer readable medium of claim 17, the method further comprising storing search patterns in a repository of search histories.

20. The computer readable medium of claim 19, the method further comprising:

determining preferred subjects of interest information for one or more the client access devices based on information stored in the repository of search histories; and

using the preferred subjects of interest information to identify newly published electronic files in one or more external information sources.

21. The computer readable medium of claim 20, the method further comprising automatically notifying the client access device of a newly published electronic file that correspond to the preferred subjects of interest.

22. The computer readable medium of claim 20, the method further comprising automatically storing the newly published electronic file in the shared information repository.

23. The computer readable medium of claim 14, wherein storing electronic files in a shared information repository further comprises allowing a client access device to associate unique metadata to one or more electronic files stored in the shared information repository.