Computer readable electronic records automated classification system
Classifying an electronic document in a computer-based system is disclosed. For each classification instance in a plurality of classification instances, a confidence data indicating a degree of confidence that the electronic document is associated with that classification instance is determined. A classification, based on a first classification instance in the plurality of classification instances, is assigned without human intervention to the electronic document if the confidence data associated with the first classification instance exceeds a first threshold.
1. A method for classifying an electronic document in a computer-based system, comprising:
determining for each classification instance in a plurality of classification instances a confidence data indicating a degree of confidence that the electronic document is associated with that classification instance; and
assigning a classification to the electronic document, without human intervention and based on a first classification instance in the plurality of classification instances, if the confidence data associated with the first classification instance exceeds a first threshold;
further comprising determining, based on a human user input, which, if any, classification should be assigned to the electronic document if the classification cannot be assigned to the document without human intervention;
wherein the classification cannot be assigned to the document without human intervention if no classification instance in the plurality of classification instances is associated with the confidence data that exceeds the first threshold; and
wherein the human user input is obtained including by presenting to a human user a graphical representation of at least two classification instances in the plurality of classification instances if each of at least a prescribed number of classification instance(s) in the plurality of classification instances has the confidence data that exceeds a second threshold that is lower than the first threshold.
2. The method recited in claim 1 , further comprising obtaining an indication of a user selection of at least one of the at least two classification instances.
3. The method recited in claim 1 , further comprising assigning a review classification to the electronic document to indicate that the electronic document is required to be reviewed by a human user if no classification instance in the plurality of classification instances is associated with the confidence data that exceeds the first threshold and fewer than a prescribed number of classification instance(s) in the plurality of classification instances has the confidence data that exceeds a second threshold.
4. The method recited in claim 1 , wherein determining the confidence data includes using information obtained by analyzing at least one training document associated with a pre-assigned classification.
5. A system for classifying an electronic document, comprising:
a processor; and
a memory coupled with the processor, wherein the memory is configured to provide the processor with instructions which when executed cause the processor to:
determine for each classification instance in a plurality of classification instances a confidence data indicating a degree of confidence that the electronic document is associated with that classification instance; and
assign a classification to the electronic document, without human intervention and based on a first classification instance in the plurality of classification instances, if the confidence data associated with the first classification instance exceeds a first threshold;
wherein the memory is further configured to provide the processor with instructions which when executed cause the processor to determine, based on a human user input, which, if any, classification should be assigned to the electronic document if the classification cannot be assigned to the document without human intervention;
wherein the classification cannot be assigned to the document without human intervention if no classification instance in the plurality of classification instances is associated with the confidence data that exceeds the first threshold; and
wherein the human user input is obtained including by presenting to a human user a graphical representation of at least two classification instances in the plurality of classification instances if each of at least a prescribed number of classification instance(s) in the plurality of classification instances has the confidence data that exceeds a second threshold that is lower than the first threshold.
6. The system recited in claim 5 , wherein obtaining an indication of a user selection of at least one of the at least two classification instances.
7. The system recited in claim 5 , wherein the memory is further configured to provide the processor with instructions which when executed cause the processor to assign a review classification to the electronic document to indicate that the electronic document is required to be reviewed by a human user if no classification instance in the plurality of classification instances is associated with the confidence data that exceeds the first threshold and fewer than a prescribed number of classification instance(s) in the plurality of classification instances has the confidence data that exceeds a second threshold.
8. The system recited in claim 5 , wherein determining the confidence data includes using information obtained by analyzing at least one training document associated with a pre-assigned classification.
9. A computer program product for classifying an electronic document, the computer program product being embodied in a non-transitory computer readable storage medium and comprising computer instructions for:
determining for each classification instance in a plurality of classification instances a confidence data indicating a degree of confidence that the electronic document is associated with that classification instance; and
assigning a classification to the electronic document, without human intervention and based on a first classification instance in the plurality of classification instances, if the confidence data associated with the first classification instance exceeds a first threshold;
further comprising computer instructions for determining, based on a human user input, which, if any, classification should be assigned to the electronic document if the classification cannot be assigned to the document without human intervention;
wherein the classification cannot be assigned to the document without human intervention if no classification instance in the plurality of classification instances is associated with the confidence data that exceeds the first threshold; and
wherein the human user input is obtained including by presenting to a human user a graphical representation of at least two classification instances in the plurality of classification instances if each of at least a prescribed number of classification instance(s) in the plurality of classification instances has the confidence data that exceeds a second threshold that is lower than the first threshold.
10. The computer program product recited in claim 9 , further comprising computer instructions for obtaining an indication of a user selection of at least one of the at least two classification instances.
11. The computer program product recited in claim 9 , wherein determining the confidence data includes using information obtained by analyzing at least one training document associated with a pre-assigned classification.