Techniques for context-based call routing for support services
Techniques are described herein for providing for automatic routing of a call to an appropriate support unit. Such techniques may comprise receiving, from an automatic call distributor (ACD) component, first information about a call that is received at a call center from a user. The techniques may be performed at a routing engine based on information about the call. Upon receiving such a call, second information associated with the call may be retrieved and, based on the first information and/or the second information associated with the call, a set of potential call reasons may be generated. A selected call reason may then be determined from the set of potential call reasons, a support unit associated with the selected call reason may be identified, and an indication of the support unit may be provided to the ACD component.
1 . A method comprising:
receive, at a routing engine from an automatic call distributor (ACD) component, first information associated with a call received from a user;
generating, by the routing engine based at least in part on the first information associated with the call, a set of potential call reasons, wherein the set of potential call reasons is generated using a trained machine-learning model;
determining, by the routing engine from the set of potential call reasons, a selected call reason;
identifying, by the routing engine, a support unit associated with the selected call reason;
providing, by the routing engine to the ACD component, an indication of the support unit; and
upon determining that the call is rerouted from the support unit to a second support unit, storing information about the rerouted call in a reroute database, and updating the trained machine-learning model based on the stored information about the rerouting of the call to the second support unit, wherein the updated trained machine-learning model is configured to correlate information about subsequent calls to the second support unit based on the stored reroute information.
2 . The method of claim 1 , further comprising retrieving, by the routing engine, second information associated with the call, wherein the second information associated with the call comprises data maintained in relation to the user.
3 . The method of claim 2 , wherein the data maintained in relation to the user comprises at least one of billing data for the user, profile data for the user, or usage data for the user.
4 . The method of claim 1 , further comprising retrieving, by the routing engine, second information associated with the call, wherein the second information associated with the call comprises data maintained in relation to a service associated with the ACD component.
5 . The method of claim 4 , wherein the data maintained in relation to a service comprises at least one of known issue data for the service or network coverage data for the service.
6 . The method of claim 1 , wherein the method further comprises assigning a weight value to individual call reasons in the set of potential call reasons, and wherein the selected call reason is determined based on the weight value.
7 . The method of claim 1 , wherein the support unit is identified based on a mapping between the support unit and the selected call reason.
8 . The method of claim 1 , further comprising identifying the user based on identifying information received in relation to the call.
9 . The method of claim 1 , wherein the set of potential call reasons is generated based on differences identified from second information.
10 . A computing device comprising:
one or more processors; and
one or more non-transitory computer-readable media storing computer-executable instructions that, when executed by the one or more processors, cause computing device to perform operations comprising:
receiving, from an automatic call distributor (ACD) component, first information associated with a call received from a user;
generating, based on the first information associated with the call, a set of potential call reasons wherein the set of potential call reasons is generated using a trained machine-learning model;
determining, from the set of potential call reasons, a selected call reason;
identifying a support unit associated with the selected call reason;
providing, to the ACD component, an indication of the support unit; and
upon determining that the call is rerouted from the support unit to a second support unit, storing information about the rerouted call in a reroute database, and updating the trained machine-learning model based on the stored information about the rerouting of the call to the second support unit, wherein the updated trained machine-learning model is configured to correlate information about subsequent calls to the second support unit based on the stored reroute information.
11 . The computing device of claim 10 , wherein the operations further comprise retrieving second information associated with the call that includes information stored in relation to the user.
12 . The computing device of claim 10 , wherein the operations further comprise retrieving second information associated with the call that includes information stored in relation to a service associated with the computing device.
13 . The computing device of claim 10 , wherein at least one potential call reason of the set of call reasons is identified based on a difference in value between a recent value and a historic average value within second information.
14 . The computing device of claim 13 , wherein the at least one potential call reason of the set of call reasons is identified upon determining that the difference in value is greater than a threshold difference.
15 . One or more non-transitory computer-readable media storing computer-executable instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:
receiving first information about a call received from a user;
retrieving second information associated with the call;
generating, based on the second information associated with the call, a set of potential call reasons, wherein the set of potential call reasons is generated using a trained machine-learning model;
determining, from the set of potential call reasons, a selected call reason;
identifying a support unit associated with the selected call reason;
providing an indication of the support unit in response to the received information; and
upon determining that the call is rerouted from the support unit to a second support unit, storing information about the rerouted call in a reroute database, and updating the trained machine-learning model based on the stored information about the rerouting of the call to the second support unit, wherein the updated trained machine-learning model is configured to correlate information about subsequent calls to the second support unit based on the stored reroute information.
16 . The one or more non-transitory computer-readable media of claim 15 , wherein the operations further comprise retrieving second information associated with the call that includes information stored in relation to the user.
17 . The one or more non-transitory computer-readable media of claim 15 , wherein the operations further comprise retrieving second information associated with the call that includes information stored in relation to a service associated with the computer-readable media, wherein the information stored in relation to the service comprises service-related operational data maintained by a service provider system and associated with a service utilized by the user.
18 . The one or more non-transitory computer-readable media of claim 15 , wherein the operations further comprise assigning a weight value to individual call reasons in the set of potential call reasons, and wherein the selected call reason is determined based on the weight value.