Reconfigurable model for auto-classification system and method
A reconfigurable automatic document-classification system and method provides classification metrics to a user and enables the user to reconfigure the classification model. The user can refine the classification model by adding or removing exemplars, creating, editing or deleting rules, or performing other such adjustments to the classification model. This technology enhances the overall transparency and defensibility of the auto-classification process.
1 . A method, comprising:
displaying, by a computer, a model reconfiguration screen that includes a metrics display pane and user interface elements, wherein the metrics display pane displays metrics for a content classification model, the metrics representing a level of accuracy of the content classification model, wherein the user interface elements are configured for making a change to example documents in the content classification model, making a change to rules for assigning classifications to documents, or a combination thereof, and wherein the computer is operable to compare the documents that are to be automatically classified with the example documents in the content classification model and automatically assigns a classification based on the rules;
receiving, by the computer through one of the user interface elements, an instruction to remove an example document from, the content classification model; and
responsive to the instruction, removing, by the computer, the example document from the content classification model, wherein the removing results in the content classification model being reconfigured so as to improve the level of accuracy.
2 . The method according to claim 1 , wherein the displaying is performed in response to the level of accuracy falling below a predetermined threshold.
3 . The method according to claim 1 , wherein the model reconfiguration screen further comprises recommended actions for optimizing the accuracy of the content classification model.
4 . The method according to claim 3 , wherein the recommended actions are updated each time the content classification model is run.
5 . The method according to claim 1 , further comprising:
running a classification test on non-classified documents from a content server;
displaying results from the classification test, the results including classifications assigned by the content classification model;
receiving, through one of the user interface elements, a rejection of one of the classifications assigned by the content classification model for a misclassified document;
receiving an expected classification for the misclassified document; and
correcting the misclassified document using the expected classification.
6 . The method according to claim 1 , wherein each of the rules specifies a rule name and a rule priority.
7 . The method according to claim 6 , wherein each of the rules further specifies a confidence level representing a level of certainty that a document matching the classification is found.
8 . A system, comprising:
a processor;
a non-transitory computer-readable medium; and
instructions stored on the non-transitory computer-readable medium and translatable by the processor for:
displaying a model reconfiguration screen that includes a metrics display pane and user interface elements, wherein the metrics display pane displays metrics for a content classification model, the metrics representing a level of accuracy of the content classification model, wherein the user interface elements are configured for making a change to example documents in the content classification model, making a change to rules for assigning classifications to documents, or a combination thereof, and wherein the computer is operable to compare the documents that are to be automatically classified with the example documents in the content classification model and automatically assigns a classification based on the rules;
receiving, through one of the user interface elements, an instruction to remove an example document from, the content classification model; and
responsive to the instruction, removing the example document from the content classification model, wherein the removing results in the content classification model being reconfigured so as to improve the level of accuracy.
9 . The system of claim 8 , wherein the displaying is performed in response to the level of accuracy falling below a predetermined threshold.
10 . The system of claim 8 , wherein the model reconfiguration screen further comprises recommended actions for optimizing the accuracy of the content classification model.
11 . The system of claim 10 , wherein the recommended actions are updated each time the content classification model is run.
12 . The system of claim 8 , wherein the instructions are further translatable by the processor for:
running a classification test on non-classified documents from a content server;
displaying results from the classification test, the results including classifications assigned by the content classification model;
receiving, through one of the user interface elements, a rejection of one of the classifications assigned by the content classification model for a misclassified document;
receiving an expected classification for the misclassified document; and
correcting the misclassified document using the expected classification.
13 . The system of claim 8 , wherein each of the rules specifies a rule name and a rule priority.
14 . The system of claim 8 , wherein each of the rules further specifies a confidence level representing a level of certainty that a document matching the classification is found.
15 . A computer program product comprising a non-transitory computer-readable medium storing instructions translatable by a processor for:
displaying a model reconfiguration screen that includes a metrics display pane and user interface elements, wherein the metrics display pane displays metrics for a content classification model, the metrics representing a level of accuracy of the content classification model, wherein the user interface elements are configured for making a change to example documents in the content classification model, making a change to rules for assigning classifications to documents, or a combination thereof, and wherein the computer is operable to compare the documents that are to be automatically classified with the example documents in the content classification model and automatically assigns a classification based on the rules;
receiving, through one of the user interface elements, an instruction to remove an example document from, the content classification model; and
responsive to the instruction, removing the example document from the content classification model, wherein the removing results in the content classification model being reconfigured so as to improve the level of accuracy.
16 . The computer program product of claim 15 , wherein the displaying is performed in response to the level of accuracy falling below a predetermined threshold.
17 . The computer program product of claim 15 , wherein the model reconfiguration screen further comprises recommended actions for optimizing the accuracy of the content classification model.
18 . The computer program product of claim 17 , wherein the recommended actions are updated each time the content classification model is run.
19 . The computer program product of claim 15 , wherein the instructions are further translatable by the processor for:
running a classification test on non-classified documents from a content server;
displaying results from the classification test, the results including classifications assigned by the content classification model;
receiving, through one of the user interface elements, a rejection of one of the classifications assigned by the content classification model for a misclassified document;
receiving an expected classification for the misclassified document; and
correcting the misclassified document using the expected classification.
20 . The computer program product of claim 15 , wherein each of the rules specifies a rule name and a rule priority.