Reconfigurable model for auto-classification system and method

1. A computer-implemented method for assessing accuracy of a reconfigurable content classification model, the method comprising:

modifying, by a computer, the reconfigurable content classification model, the modifying including importing from a document source into a container of the reconfigurable content classification model, example documents that have been identified as exemplars for a content classification;

randomly selecting, from the example documents imported into the container, a set of test documents for testing the reconfigurable content classification model for accuracy;

running, by the computer using the reconfigurable content classification model, a test classification on the set of test documents;

computing, by the computer, performance metrics for the reconfigurable content classification model based on the test classification, the performance metrics representing a level of accuracy of the reconfigurable content classification model after running the test classification on the set of test documents;

presenting, by the computer through a model reconfiguration user interface, the performance metrics for the reconfigurable content classification model and a confidence level for each respective test document of the set of test documents, wherein the confidence level refers to a level of certainty that the respective test document is correctly classified using the reconfigurable content classification model, the model reconfiguration user interface having user interface elements wherein reviewing the set of test documents presented with confidence levels enables a user to locate any error in the test classification; and

responsive to an instruction received through the model reconfiguration user interface to add or remove an example document from the container adjusting or reconfiguring, by the computer, the reconfigurable content classification model accordingly so as to improve the level of accuracy.

2. The computer-implemented method according to claim 1 , wherein the user interface elements further include a user interface element for viewing, adding, or deleting a classification rule applicable to the content classification.

3. The computer-implemented method according to claim 2 , wherein the classification rule is one of a plurality of classification rules for the reconfigurable content classification model, each respective rule of the plurality of classification rules having:

a rule priority that determines an order in which the respective rule is applied;

a confidence level to be applied to a document when the document satisfies a condition specified by the respective rule; and

an applied classification that is to be applied to the document.

4. The computer-implemented method according to claim 1 , wherein the importing further includes importing additional example documents that have been identified as exemplars for a plurality of content classifications in a hierarchical classification scheme.

5. The computer-implemented method according to claim 1 , further comprising:

classifying, using the reconfigurable content classification model, groups of documents, entire databases, entire drives, subsets thereof, or individually selected groups of documents.

6. The computer-implemented method according to claim 5 , further comprising:

after the classifying is complete or as the classifying is being performed, computing and displaying metrics for at least one of: a number of documents processed, a number of documents that have been classified, a number of documents that have not been unclassified, a number of documents that have been rejected, or a number of documents that have been assigned a confidence level.

7. The computer-implemented method according to claim 1 , wherein the document source comprises a content server.

8. A system for assessing accuracy of a reconfigurable content classification model, the system comprising:

a processor;

a non-transitory computer-readable medium; and

stored instructions translatable by the processor for:

modifying the reconfigurable content classification model, the modifying including importing from a document source into a container of the reconfigurable content classification model, example documents that have been identified as exemplars for a content classification;

randomly selecting, from the example documents imported into the container, a set of test documents for testing the reconfigurable content classification model for accuracy;

running, by the computer using the reconfigurable content classification model, a test classification on the set of test documents;

computing performance metrics for the reconfigurable content classification model based on the test classification, the performance metrics representing a level of accuracy of the reconfigurable content classification model after running the test classification on the set of test documents;

presenting, through a model reconfiguration user interface, the performance metrics for the reconfigurable content classification model and a confidence level for each respective test document of the set of test documents, wherein the confidence level refers to a level of certainty that the respective test document is correctly classified using the reconfigurable content classification model, the model reconfiguration user interface having user interface elements wherein reviewing the set of test documents presented with confidence levels enables a user to locate any error in the test classification; and

responsive to an instruction received through the model reconfiguration user interface to add or remove an example document from the container adjusting or reconfiguring the reconfigurable content classification model accordingly so as to improve the level of accuracy.

9. The system of claim 8 , wherein the user interface elements further include a user interface element for viewing, adding, or deleting a classification rule applicable to the content classification.

10. The system of claim 9 , wherein the classification rule is one of a plurality of classification rules for the reconfigurable content classification model, each respective rule of the plurality of classification rules having:

a rule priority that determines an order in which the respective rule is applied;

a confidence level to be applied to a document when the document satisfies a condition specified by the respective rule; and

an applied classification that is to be applied to the document.

11. The system of claim 8 , wherein the importing further includes importing additional example documents that have been identified as exemplars for a plurality of content classifications in a hierarchical classification scheme.

12. The system of claim 8 , wherein the stored instructions are further translatable by the processor for:

classifying, using the reconfigurable content classification model, groups of documents, entire databases, entire drives, subsets thereof, or individually selected groups of documents.

13. The system of claim 12 , wherein the stored instructions are further translatable by the processor for:

after the classifying is complete or as the classifying is being performed, computing and displaying metrics for at least one of: a number of documents processed, a number of documents that have been classified, a number of documents that have not been unclassified, a number of documents that have been rejected, or a number of documents that have been assigned a confidence level.

14. The system of claim 8 , wherein the document source comprises a content server.

15. A computer program product for assessing accuracy of a reconfigurable content classification model, the computer program product comprising a non-transitory computer-readable medium storing instructions translatable by a processor for:

modifying the reconfigurable content classification model, the modifying including importing, from a document source into a container of the reconfigurable content classification model, example documents that have been identified as exemplars for a content classification;

randomly selecting, from the example documents imported into the container, a set of test documents for testing the reconfigurable content classification model for accuracy;

running, by the computer using the reconfigurable content classification model, a test classification on the set of test documents;

computing performance metrics for the reconfigurable content classification model based on the test classification, the performance metrics representing a level of level of accuracy of the reconfigurable content classification model after running the test classification on the set of test documents;

presenting, through a model reconfiguration user interface, the performance metrics for the reconfigurable content classification model and a confidence level for each respective test document of the set of test documents, wherein the confidence level refers to a level of certainty that the respective test document is correctly classified using the reconfigurable content classification model, the model reconfiguration user interface having user interface elements wherein reviewing the set of test documents presented with confidence levels enables a user to locate any error in the test classification; and

responsive to an instruction received through the model reconfiguration user interface to add or remove an example document from the container, adjusting or reconfiguring the reconfigurable content classification model accordingly so as to improve the level of accuracy.

16. The computer program product of claim 15 , wherein the user interface elements further include a user interface element for viewing, adding, or deleting a classification rule applicable to the content classification.

17. The computer program product of claim 16 , wherein the classification rule is one of a plurality of classification rules for the reconfigurable content classification model, each respective rule of the plurality of classification rules having:

a rule priority that determines an order in which the respective rule is applied;

a confidence level to be applied to a document when the document satisfies a condition specified by the respective rule; and

an applied classification that is to be applied to the document.

18. The computer program product of claim 15 , wherein the importing further includes importing additional example documents that have been identified as exemplars for a plurality of content classifications in a hierarchical classification scheme.

19. The computer program product of claim 15 , wherein the instructions are further translatable by the processor for:

classifying, using the reconfigurable content classification model, groups of documents, entire databases, entire drives, subsets thereof, or individually selected groups of documents.

20. The computer program product of claim 19 , wherein the instructions are further translatable by the processor for:

after the classifying is complete or as the classifying is being performed, computing and displaying metrics for at least one of: a number of documents processed, a number of documents that have been classified, a number of documents that have not been unclassified, a number of documents that have been rejected, or a number of documents that have been assigned a confidence level.