Systems and methods to terminate an active communication
Contact centers often provide a restricted amount of time for agents to perform post-call activities, including identifying and selecting a relevant completion code from a large set of codes. Providing automated systems and methods to identify the relevant code or codes can present a limited number of the codes (e.g., quick release codes) to the agent and/or automatically enter the completion code. When the quick release code is entered, the record of the call is then updated with the code and the communication is terminated as a single step.
1. A system, comprising:
a data storage;
a network interface to a network;
a processor configured with machine-readable instructions; and
wherein the processor:
monitors, by the network interface, a communication between a customer utilizing a customer communication device and an agent utilizing an agent communication device conducted over the network;
determine, from the communication, attributes of the communication;
cause a number of indicia of the attributes of the communication to be presented for selection of one thereof, by an input to the agent communication device; and
in response to receiving the input, alone and without any subsequent inputs, selecting the one of the number of indicia, both terminating the communication and updating a record in the data storage to comprise the selected one of the number of indicia.
2. The system of claim 1 , wherein at least one of the attributes of the communication describes at least one of a reason for the terminating, an attribute of the customer, a content of the communication.
3. The system of claim 1 , wherein the processor determines attributes of the communication comprising providing the communication to a neural network trained to determine attributes from the communication.
4. The system of claim 3 , wherein neural network is trained to determine the attributes of the communication, comprising:
collecting a set of communications from a database maintained in the data storage;
applying one or more transformations to each communication in the collected set of communications including substituting a portion of content of the communication with a different content having a same meaning as the portion of the content to create a modified set of communications;
creating a first training set comprising the collected set of communications, the modified set of communications, and a set of communications absent the attributes of the communication;
training the neural network in a first state using the first training set;
creating a second training set for a second state of training comprising the first training set and the collected set of communications absent the attributes of the communication that are incorrectly detected as having the attributes of the communication after the first state of training; and
training the neural network in a second state using the second training set.
5. The system of claim 1 , wherein the processor causes the number of indicia of the attributes of the communication to be presented for selection, further comprising the processor performing:
determining a reliability score for each of the attributes of the communication;
accessing a threshold reliability score; and
causing the number of indicia of the attributes of the communication to be presented for selection for those attributes of the communication having a reliability score of at least the threshold reliability score.
6. The system of claim 5 , wherein the processor determines the reliability score as a degree of match to a model communication content having at least one of the attributes of the communication.
7. The system of claim 5 , wherein a degree of match is determined to exceed a conclusion threshold and, in response, terminating the communication prior to receiving the input.
8. The system of claim 1 , wherein the communication comprises a real-time communication and termination of the communication comprises terminating media exchanged between the agent communication device and the customer communication device.
9. The system of claim 1 , wherein the communication comprises a turn-based communication comprising a number of messages and termination of the communication comprises causing signaling a communication server to route a message received from the customer communication device, which is received after the termination, to a destination other than the agent communication device.
10. A method, comprising:
monitoring a communication between a customer utilizing a customer communication device and an agent utilizing an agent communication device conducted over a network;
determining, from the communication, attributes of the communication;
causing a number of indicia of the attributes of the communication to be presented for selection of one thereof, by an input to the agent communication device; and
in response to receiving the input, alone and without any subsequent inputs, selecting the one of the number of indicia, terminating the communication and updating a record in data storage to comprise the selected one of the number of indicia.
11. The method of claim 10 , wherein at least one of the attributes of the communication describes at least one of a reason for the terminating, an attribute of the customer, a content of the communication.
12. The method of claim 10 , wherein determining the attributes of the communication comprising providing the communication to a neural network trained to determine attributes from the communication.
13. The method of claim 12 , wherein neural network is trained to determine the attributes of the communication, comprising:
collecting a set of communications from a database maintained in the data storage;
applying one or more transformations to each communication in the collected set of communications including substituting a portion of content of the communication with a different content having a same meaning as the portion of the content to create a modified set of communications;
creating a first training set comprising the collected set of communications, the modified set of communications, and a set of communications absent the attributes of the communication;
training the neural network in a first state using the first training set;
creating a second training set for a second state of training comprising the first training set and the collected set of communications absent the attributes of the communication that are incorrectly detected as having the attributes of the communication after the first state of training; and
training the neural network in a second state using the second training set.
14. The method of claim 10 , wherein causing the number of indicia of the attributes to be presented for selection, further comprises:
determining a reliability score for each of the attributes;
accessing a threshold reliability score; and
causing the number of indicia of the attributes to be presented for selection for those attributes having a reliability score of at least the threshold reliability score.
15. The method of claim 14 , wherein determining the reliability score as a degree of match to a model communication content having at least one of the attributes.
16. The method of claim 15 , wherein the degree of match is determined to exceed a conclusion threshold and, in response, terminating the communication prior to receiving the input.
17. The method of claim 10 , wherein the communication comprises a real-time communication and termination of the communication comprises terminating media exchanged between the agent communication device and the customer communication device.
18. The method of claim 10 , wherein the communication comprises a turn-based communication comprising a number of messages and termination of the communication comprises causing signaling a communication server to route a message received from the customer communication device, which is received after the termination, to a destination other than the agent communication device.
19. A system, comprising:
means for monitoring a communication between a customer utilizing a customer communication device and an agent utilizing an agent communication device conducted over a network;
means for determining, from the communication, attributes of the communication;
means for causing a number of indicia of the attributes of the communication to be presented for selection of one thereof, by an input to the agent communication device; and
means for, in response to receiving the input, alone and without any subsequent inputs, selecting the one of the number of indicia, terminating the communication and updating a record in data storage to comprise the selected one of the number of indicia.
20. The system of claim 19 , wherein at least one of the attributes of the communication describes at least one of a reason for the terminating, an attribute of the customer, a content of the communication.