SYSTEMS AND METHODS FOR ORGANIZING DATA SETS

• US 20090228499A1
• Filed: 03/05/2008
• Published: 09/10/2009
• Est. Priority Date: 03/05/2008
• Status: Active Grant

• Current Assignee: Kofax Incorporated
• Original Assignee: Kofax Image Products Incorporated (Kofax Ltd.)
• Inventors: Sarah, Anthony, Schmidtler, Mauritius A.R., Thompson, Stephen Michael, Amtrup, Jan W.
• Granted Patent: US 9,082,080 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G06F 16/22 Indexing; Data structures therefor; Storage structures
• CPC Class Codes: G06F 16/24578 using ranking
• CPC Class Codes: G06F 16/285 Clustering or classification
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 20/10 using kernel methods, e.g. support vector machines [SVM]

Abstract

A method is provided for organizing data sets. In use, an automatic decision system is created or updated for determining whether data elements fit a predefined organization or not, where the decision system is based on a set of preorganized data elements. A plurality of data elements is organized using the decision system. At least one organized data element is selected for output to a user based on a score or confidence from the decision system for the at feast one organized data element. Additionally, at least a portion of the at least one organized data element is output to the user. A response is received from the user comprising at least one of a confirmation, modification, and a negation of the organization of the at least one organized data element. The automatic decision system is recreated or updated based on the user response. Other embodiments are also presented.

Claims

1. A method for organizing data sets, comprising:

creating or updating an automatic decision system for determining whether data elements fit a predefined organization or not, the decision system being based on a set of preorganized data elements;

organizing a plurality of data elements using the decision system;

selecting at least one organized data element for output to a user based on a score or confidence from the decision system of the at least one organized data element;

outputting at least a portion of the at least one organized data element to the user;

receiving a response from the user comprising at least one of a confirmation, modification, and a negation of the organization of the at least one organized data element; and

recreating or updating the automatic decision system based on the user response,

2. The method of claim 1, wherein creating or updating the automatic decision system is based on an organization provided in the set of preorganized data elements.

3. The method of claim 1, wherein the organization for creating or updating the automatic decision system is based on a clustering of an originally unorganized data set.

4. The method of claim 1, wherein the at least one organized data element is selected for output based on having a decision score or confidence above a predetermined value.

5. The method of claim 1, wherein the at least one organized data element is selected for output based on having a decision score or confidence between predetermined values.

6. The method of claim 1, wherein the at least one organized data element is selected for output based on having a decision score or confidence indicating that a process of organizing the data element was more difficult relative to one or more of the other data elements.

7. The method of claim 1, wherein the decision system is a binary classifier.

8. The method of claim 1, wherein if the user response is a confirmation, the at least one organized data element is used as a positive example for recreating or updating the decision system;

wherein if the user response is a negation or modification, the at least one organized data element is used as a negative example for recreating or updating the decision system.

9. The method of claim 1, further comprising reorganizing at least a portion of the plurality of data elements after recreating or updating the decision system.

10. The method of claim 1, further comprising organizing a new plurality of data elements after recreating or updating the decision system.

11. The method of claim 1, where the data elements are documents and the organization is given by an assignment of documents to categories based on the content of the documents.

12. The method of claim L where the data elements are elements within a document, and the organization is given by an assignment of labels to said elements based on the function of said elements in said document.

13. A method for detecting limits to the quality in a dynamic organization process, comprising:

receiving user input affirming, negating, or modifying organization of a plurality of data elements during or after an organization process;

determining a quality of the organization based on the user input;

determining a stability in decisions made during the organization of the plurality of data elements;

monitoring the quality and the stability;

determining that the organization has reached a limit concerning at least one of the quality and stability of the organization by evaluating the at least one of the quality and stability of the organization over a period of time or number of cycles; and

outputting an indication of the determination to at least one of a user, a system and another process.

14. The method of claim 13, wherein the determination is that the quality of the organization is not likely to significantly change if the at least one of the quality and stability do not change by more than a predetermined amount over a given amount of time.

15. The method of claim 13, wherein the determination is that the organization process is having difficulty in handling the data if the at least one of the quality and stability oscillate over time.

16. The method of claim 13, further comprising outputting at least portions of the data elements to the user during the organization process, wherein the at least one of the quality and the stability is updated each time user input is received.

17. The method of claim 13, further comprising changing an input or its representation used during the organization upon determining that the organization process has reached a limit.

18. The method of claim 13, further comprising implementing different decision functions than currently being used for organization upon determining that the organization process has reached a limit.

19. The method of claim 13, further comprising altering the organization of data elements upon determining that the organization process has reached a limit.

20. The method of claim 19, wherein altering the organization of data elements includes combining data elements from at least two aspects of the organization into a new single aspect.

21. The method of claim 20, further comprising, in a second organization process, separating the data elements in the new single aspect back into the at least two aspects.

22. The method of claim 13, where the data elements are documents and the organization is based on an assignment of documents to categories based on the content of the documents.

23. The method of claim 13, where the data elements are elements within a document, and the organization is based on an assignment of labels to said elements based on the function of said elements in said document.

24. A computer program product comprising a computer readable medium having computer code thereon, which when executed on a computer causes the computer to:

create or update an automatic decision system for determining whether data elements fit a predefined organization or not, the decision system being based on a set of preorganized data elements;

organize a plurality of data elements using the decision system;

select at least one organized data element for output to a user based on a score or confidence from the decision system of the at least one organized data element;

output at least a portion of the at least one organized data element to the user;

receive a response from the user comprising at least one of a confirmation, modification, and a negation of the organization of the at least one organized data element; and

recreate or update the automatic decision system based on the user response.

25. A computer program product comprising a computer readable medium having computer code thereon, which when executed on a computer causes the computer to;

receive user input affirming, negating, or modifying organization of a plurality of data elements during or after an organization process;

determine a quality of the organization based on the user input;

determine a stability in decisions made during the organization of the plurality of data elements;

monitor the quality and the stability;

determine that the organization has reached a limit concerning at least one of the quality and stability of the organization by evaluating the at least one of the quality and stability of the organization over a period of time or number of cycles; and

output an indication of the determination to at least one of a user, a system and another process.