System and method for automatic quality management in a contact center environment
A method for automatically managing a recorded interaction between a customer and an agent of a contact center includes: extracting, by a processor, features from the recorded interaction; computing, by the processor, a score of the recorded interaction by supplying the features to a prediction model; detecting, by the processor, a condition based on the score; matching, by the processor, the condition with an action; and transmitting, by the processor, a command to control a component of the contact center to perform the action.
1. A method for automatically managing a recorded interaction between a customer and an agent of a contact center, the method comprising:
extracting, by a processor, features from the recorded interaction;
computing, by the processor, an agent performance evaluation score of the recorded interaction by supplying the features to a prediction model;
detecting, by the processor, a condition based on the agent performance evaluation score;
matching, by the processor, the condition with an action; and
transmitting, by the processor, a command to control a component of the contact center to perform the action on the recorded interaction in accordance with the agent performance evaluation score of the recorded interaction.
2. The method of claim 1 , wherein the recorded interaction comprises content and metadata, and
wherein the features comprise at least one of:
keywords detected in the content of the recorded interaction;
one or more topics in the content of the recorded interaction;
a length of the recorded interaction;
a number of agent transfers during the recorded interaction; and
a net promoter score associated with the recorded interaction.
3. The method of claim 1 , wherein the prediction model is trained by:
extracting, by the processor, a plurality of historical interaction features from a plurality of historical interactions;
receiving, by the processor, a plurality of manual evaluations, each of the manual evaluations corresponding to one of the plurality of historical interactions, and each of the manual evaluations comprising a score; and
training, by the processor, the prediction model in accordance with a relationship between the historical interaction features of the historical interactions and the scores of corresponding ones of the manual evaluations.
4. The method of claim 1 , wherein the condition is satisfied when the agent performance evaluation score is within a threshold range corresponding to compliance with agent performance standards, and
wherein the component of the contact center is a call recording server.
5. The method of claim 4 , wherein the command is configured to control the call recording server to delete the recorded interaction.
6. The method of claim 4 , wherein the command is configured to control the call recording server to set an expiration date on the recorded interaction.
7. The method of claim 1 , wherein the condition is satisfied when the agent performance evaluation score is within a threshold range corresponding to not in compliance with agent performance standards, and
wherein the component of the contact center is a quality management server.
8. The method of claim 7 , wherein the command is configured to control the quality management server to send an alert to a human supervisor.
9. The method of claim 7 , wherein the command is configured to control the quality management server to assign an evaluation task to a human supervisor.
10. A system for automatically managing a recorded interaction between a customer and an agent of a contact center comprises:
a processor; and
a memory storing instructions that, when executed by the processor, cause the processor to:
extract features from the recorded interaction;
compute an agent performance evaluation score of the recorded interaction by supplying the features to a prediction model;
detect a condition based on the agent performance evaluation score;
match the condition with an action; and
transmit a command to control a component of the contact center to perform the action on the recorded interaction in accordance with the agent performance evaluation score of the recorded interaction.
11. The system of claim 10 , wherein the recorded interaction comprises content and metadata, and
wherein the features comprise at least one of:
keywords detected in the content of the recorded interaction;
one or more topics in the content of the recorded interaction;
a length of the recorded interaction;
a number of agent transfers during the recorded interaction; and
a net promoter score associated with the recorded interaction.
12. The system of claim 10 , wherein the prediction model is trained by:
extracting a plurality of historical interaction features from a plurality of historical interactions;
receiving a plurality of manual evaluations, each of the manual evaluations corresponding to one of the plurality of historical interactions, and each of the manual evaluations comprising a score; and
training the prediction model in accordance with a relationship between the historical interaction features of the historical interactions and the scores of corresponding ones of the manual evaluations.
13. The system of claim 10 , wherein the condition is satisfied when the agent performance evaluation score is within a threshold range corresponding to compliance with agent performance standards, and
wherein the component of the contact center is a call recording server.
14. The system claim 13 , wherein the command is configured to control the call recording server to delete the recorded interaction.
15. The system of claim 13 , wherein the command is configured to control a call recording server to set an expiration date on the recorded interaction.
16. The system of claim 10 , wherein the condition is satisfied when the agent performance evaluation score is within a threshold range corresponding to not in compliance with agent performance standards, and
wherein the component of the contact center is a quality management server.
17. The system of claim 16 , wherein the command is configured to control the quality management server to send an alert to a human supervisor.
18. The system of claim 16 , wherein the command is configured to control the quality management server to assign an evaluation task to a human supervisor to perform a manual evaluation of the recorded interaction.
19. A system for automatically managing a recorded interaction between a customer and an agent of a contact center, the system comprising:
means for extracting features from the recorded interaction;
means for computing an agent performance evaluation score of the recorded interaction by supplying the features to a prediction model;
means for detecting a condition based on the agent performance evaluation score;
means for matching the condition with an action; and
means for transmitting a command to control a component of the contact center to perform the action on the recorded interaction in accordance with the agent performance evaluation score of the recorded interaction.
20. The system of claim 19 , wherein the recorded interaction comprises content and metadata, and
wherein the features comprise at least one of:
keywords detected in the content of the recorded interaction;
one or more topics in the content of the recorded interaction;
a length of the recorded interaction;
a number of agent transfers during the recorded interaction; and
a net promoter score associated with the recorded interaction.