Systems and methods for organizing data sets

1. A computer-implemented method for confirming or rejecting a most relevant example, the method comprising:

generating a binary decision model by training a binary classifier using a plurality of training documents;

classifying one or more test documents into one of a plurality of document type categories using the binary decision model, wherein the one or more test documents lack a user-defined category label;

selecting a most relevant example of the classified test documents from among the classified test documents, wherein the most relevant example of the classified test documents is the test document having a classification score closest to a boundary between a positive decision and a negative decision concerning the test document belonging to a particular one of the plurality of categories;

displaying, using a display of the computer, the most relevant example of the classified test documents to a user;

receiving, via the computer and from the user, a confirmation or a negation of a classification label of the most relevant example of the classified test documents; and

storing the confirmation or the negation of the classification label of the most relevant example of the classified test documents to a memory of the computer.

2. The method of claim 1 , wherein the plurality of training documents comprises a plurality of positive examples of documents belonging to a particular one of the plurality of categories, and a plurality of negative examples of documents not belonging to the particular one of the plurality of categories.

3. The method of claim 1 , wherein the only response options for the user to provide, in response to reviewing the displayed most relevant example, are the confirmation and the negation.

4. The method of claim 1 , further comprising generating a second binary decision model by training the binary classifier using the plurality of training documents and the confirmation or the negation of the classification label of the most relevant example of the classified test documents.

5. The method of claim 4 , wherein the second binary decision model is generated immediately after receiving the confirmation or the negation from the user.

6. The method of claim 4 , wherein the confirmation or the negation of the classification label of the most relevant example of the classified test documents is the single example of user input used in generating the second binary decision model.

7. The method of claim 4 , further comprising reclassifying the one or more test documents into one of a plurality of categories using the second binary decision model.

8. The method of claim 7 , further comprising:

selecting a most relevant example of the reclassified test documents from among the reclassified test documents;

displaying, using the display of the computer, the most relevant example of the reclassified test documents to the user;

receiving, via the computer and from the user, a confirmation or a negation of a classification label of the most relevant example of the reclassified test documents; and

storing the confirmation or the negation of the classification label of the most relevant example of the reclassified test documents to the memory of the computer; and

generating a new binary decision model by training the binary classifier using the plurality of training documents and the confirmation or the negation of the classification label of the most relevant example of the reclassified test documents.

9. The method of claim 8 , further comprising iteratively repeating, in a loop, the reclassifying, the selecting, the displaying, the receiving, the storing; and

the generating; and

terminating the iterative loop in response to receiving an indication that sufficient relevant examples have been labeled.

10. A computer program product, comprising a non-transitory computer readable storage medium having program instructions embodied therewith, the program instructions executable by a computer to perform a method comprising:

generating, using a processor of the computer, a binary decision model by training a binary classifier using a plurality of training documents;

classifying, using the processor, one or more test documents into one of a plurality of document type categories using the binary decision model, wherein the one or more test documents lack a user-defined category label;

selecting, using the processor, a most relevant example of the classified test documents from among the classified test documents, wherein the most relevant example of the classified test documents is the test document having a classification score closest to a boundary between a positive decision and a negative decision concerning the test document belonging to a particular one of the plurality of categories;

displaying, using a display of the computer, the most relevant example of the classified test documents to a user;

receiving, via the computer and from the user, a confirmation or a negation of a classification label of the most relevant example of the classified test documents; and

storing the confirmation or the negation of the classification label of the most relevant example of the classified test documents to a memory of the computer.

11. The computer program product of claim 10 , wherein the plurality of training documents comprises a plurality of positive examples of documents belonging to a particular one of the plurality of categories, and a plurality of negative examples of documents not belonging to the particular one of the plurality of categories.

12. The computer program product of claim 10 , wherein the only response options for the user to provide, in response to reviewing the displayed most relevant example, are the confirmation and the negation.

13. The computer program product of claim 10 , further comprising program instructions executable by a computer to generate a second binary decision model by training the binary classifier using the plurality of training documents and the confirmation or the negation of the classification label of the most relevant example of the classified test documents.

14. The computer program product of claim 13 , wherein the second binary decision model is generated immediately after receiving the confirmation or the negation from the user.

15. The computer program product of claim 13 , wherein the confirmation or the negation of the classification label of the most relevant example of the classified test documents is the single example of user input used in generating the second binary decision model.

16. The computer program product of claim 13 , further comprising program instructions executable by a computer to reclassify the one or more test documents into one of a plurality of categories using the second binary decision model.

17. The computer program product of claim 16 , further comprising program instructions executable by a computer to:

select a most relevant example of the reclassified test documents from among the reclassified test documents;

display, using the display of the computer, the most relevant example of the reclassified test documents to the user;

receive, via the computer and from the user, a confirmation or a negation of a classification label of the most relevant example of the reclassified test documents; and

store the confirmation or the negation of the classification label of the most relevant example of the reclassified test documents to the memory of the computer; and

generate a new binary decision model by training the binary classifier using the plurality of training documents and the confirmation or the negation of the classification label of the most relevant example of the reclassified test documents.

18. A computer system, comprising a processor and logic integrated with/executable by the processor to cause the system to:

generate, using the processor, a binary decision model by training a binary classifier using a plurality of training documents;

classifying, using the processor, one or more test documents into one of a plurality of document type categories using the binary decision model, wherein the one or more test documents lack a user-defined category label;

selecting, using the processor, a most relevant example of the classified test documents from among the classified test documents, wherein the most relevant example of the classified test documents is the test document having a classification score closest to a boundary between a positive decision and a negative decision concerning the test document belonging to a particular one of the plurality of categories;

display, using a display of the computer system, the most relevant example of the classified test documents to a user;

receive, via the computer system and from the user, a confirmation or a negation of a classification label of the most relevant example of the classified test documents; and

store, in the computer system, the confirmation or the negation of the classification label of the most relevant example of the classified test documents to a memory of the computer.