System and method for organizing repositories of semi-structured documents such as email

• US 6,592,627 B1
• Filed: 06/10/1999
• Issued: 07/15/2003
• Est. Priority Date: 06/10/1999
• Status: Expired due to Term

• Current Assignee: Google LLC (Alphabet Inc.)
• Original Assignee: International Business Machines Corporation
• Inventors: Srikant, Ramakrishnan, Shafer, John Christopher, Sarawagi, Sunita, Ho, Ching-Tien Howard, Gunopulos, Dimitrios, Bayardo, Roberto Javier, Agrawal, Rakesh
• Primary Examiner(s): Feild, Jospeh H.
• Assistant Examiner(s): YUAN, ALMARI ROMERO
• Application Number:
  US09/329,684
• Time in Patent Office: 1,496 Days
• Field of Search: 707/513, 707/5, 707/10
• US Class Current: 715/234
• CPC Class Codes: G06F 16/93 Document management systems

Abstract

A user can easily organize computerized document folders by associating a few sample documents in the document database with each folder. The present invention learns folder profiles based on the sample documents and moves the remaining documents into the folders accordingly. In this way, the user can construct new folders, or rearrange existing folders, or cause the computer to automatically rearrange and maintain the folders. This is particularly useful for managing a database of perhaps thousands of emails.

Claims

1. A computer including at least one computer input device and at least one computer usable media having computer usable means for organizing one or more electronically-stored folders holding one or more documents, the computer usable means having:

means for receiving, from the input device, at least one signal representative of user-selected document classification indicia;

means for determining a profile of at least one folder based on the user-selected document classification indicia;

means for determining, for each document, respective folder probabilities, each folder probability representing the probability of the document fitting the profile of the respective folder;

means for defining a destination of a document to be the folder associated with the highest folder probability; and

means for defining a confidence of a document properly being in a destination to be the ratio of the highest folder probability for the document to a second-highest folder probability for the document.

2. The computer of claim 1, wherein the user-selected document classification indicia includes at least one sample document representing a user-desired example of documents in a user-selected folder.

3. The computer of claim 2, further comprising:

means for receiving one or more folder establishing signals from the input device, the document classification indicia representing a user-desired profile of at least some folders; and

means for automatically moving one or more documents into the folders, based on the means for determining a profile.

4. The computer of claim 1, wherein the user-selected document classification indicia includes at least one of:

at least one classification rule.

5. The computer of claim 1, further comprising:

means for automatically moving one or more test documents into the folders, based on the means for determining a profile; and

means for presenting at least a portion of at least some test documents, along with the associated folders, on a computer output device associated with the computer, such that a user can observe the test documents with folders and determine the efficacy of the means for determining a profile.

6. The computer of claim 5, wherein the means for receiving at least one signal representative of document classification indicia can receive additional user-defined document classification indicia generated in response to the means for presenting, the additional user-defined document classification indicia being usable by the means for determining a profile to refine the determinations of the profiles of the folders.

7. The computer of claim 1, further comprising means for requesting additional user-defined document classification indicia.

8. The computer of claim 1, further comprising means for generating a folder, based on the user-defined document classification indicia.

9. The computer of claim 1, further comprising means for grouping documents in at least one folder into sub-folders, based on the user-defined document classification indicia.

10. The computer of claim 1, wherein the documents are email documents, and the computer further comprises means for routing incoming email documents into one or more folders, based on the user-defined document classification indicia.

11. A computer-implemented method for organizing email documents in a database into one or more folders, comprising:

receiving one or more sample email documents from the user, the sample documents being a relatively small subset of the documents in the database; and

based on the sample documents, automatically associating substantially all of the documents in the database with one or more folders, the associating act including;

determining, for each document, respective folder probabilities, each folder probability representing the probability of the document fitting the profile of the respective folder;

defining a destination of a document to be the folder associated with the highest folder probability; and

defining a confidence of a document properly being in a destination to be the ratio of the highest folder probability for the document to a second-highest folder probability for the document.

12. The method of claim 11, wherein the sample documents establish user-defined classification indicia, and the associating step includes determining a profile of at least one folder based on the user-selected document classification indicia.

13. The method of claim 12, wherein the user-selected document classification indicia includes at least one of:

at least one classification rule.

14. The method of claim 12, further comprising:

receiving one or more folder establishing signals from a computer input device, the document classification indicia representing a user-desired profile of at least some folders.

15. The method of claim 12, further comprising:

automatically moving one or more test documents into the folders, based on the determining step; and

presenting at least a portion of at least some test documents, along with the associated folders, on a computer output device, such that a user can observe the test documents with folders and determine the efficacy of the step of determining a profile.

16. The method of claim 15, further comprising:

receiving additional user-defined document classification indicia generated in response to the presenting step, the additional user-defined document classification indicia being usable for refining the determinations of the profiles of the folders.

17. The method of claim 12, further comprising requesting additional user-defined document classification indicia.

18. The method of claim 12, further comprising generating a folder, based on the user-defined document classification indicia.

19. The method of claim 12, further comprising grouping documents in at least one folder into sub-folders, based on the user-defined document classification indicia.

20. The method of claim 12, further comprising routing incoming email documents into one or more folders, based on the user-defined document classification indicia.

21. A computer program device comprising:

a computer program storage device readable by a digital processing apparatus; and

a program means on the program storage device and including instructions executable by the digital processing apparatus for performing method steps for organizing semi-structured documents into folders, the method steps comprising;

receiving, from a computer input device, at least one user-generated sample signal representing one or more sample documents;

based thereon, establishing a profile for at least two document folders, the establishing act including;

determining, for each document, respective folder probabilities, each folder probability representing the probability of the document fitting the profile of the respective folder;

defining a destination of a document to be the folder associated with the highest folder probability; and

defining a confidence of a document properly being in a destination to be the ratio of the highest folder probability for the document to a second-highest folder probability for the document; and

based on the profile, moving documents from a database into the folders.

22. The computer program device of claim 21, wherein the documents are emails.

23. The computer program device of claim 21, wherein the method steps further comprise receiving one or more folder establishing signals from the computer input device.

24. The computer program device of claim 21, wherein the method steps further comprise:

in response to the establishing step, presenting at least a portion of at least some test documents, along with the associated folders, on a computer output device, such that a user can observe the test documents with folders.

25. The computer program device of claim 24, wherein the method steps further comprise:

receiving additional user-defined document classification indicia generated in response to the presenting step, the additional user-defined document classification indicia being usable for refining the determinations of the profiles of the folders.

26. The computer program device of claim 21, wherein the method steps further comprise requesting additional sample signals from the user.

27. The computer program device of claim 21, wherein the method steps further comprise generating a folder, based on the sample signal.

28. The computer program device of claim 21, wherein the method steps further comprise grouping documents in at least one folder into sub-folders, based on the sample signal.

29. The computer program device of claim 21, wherein the method steps further comprise routing incoming documents into one or more folders, based on the sample signal.