Utilizing machine learning with call histories to determine support queue positions for support calls
A device receives, from a user device, a communication associated with a support issue encountered by a user of the user device and assigns the communication to a position in a support queue based on when the communication is received, wherein the support queue includes data identifying positions of other communications received from other users, and data identifying when the other communications were received. The device processes data identifying the communication and historical communication data describing prior communications associated with the user, with a model, to determine an average time spent on hold by the user for the prior communications. The device modifies the position of the communication in the support queue based on the average time and performs one or more actions based on modifying the position of the communication in the support queue.
1. A method, comprising:
receiving, by a device, data identifying a current communication associated with a user;
processing, by the device, the data identifying the current communication and historical communication data describing prior communications associated with the user and received prior to the current communication, with a machine learning model, to determine a score for the current communication; and
placing, by the device, the current communication in a queue based on the score for the current communication and other scores for other communications included in the queue.
2. The method of claim 1 , further comprising:
receiving other data identifying the other communications; and
processing the other data and other historical communication data describing other prior communications associated with other users, with the machine learning model, to determine the other scores.
3. The method of claim 1 , further comprising:
modifying, in the queue and based on placing the current communication in the queue, at least one position of at least one other communication of the other communications.
4. The method of claim 1 , wherein processing the data identifying the current communication and the historical communication data comprises:
determining, based on the historical communication data, a time period between the current communication and a historical communication identified by the historical communication data,
wherein the score for the current communication is based on the time period.
5. The method of claim 1 , wherein the machine learning model is associated with at least one factor used to determine the score for the current communication,
the at least one factor including at least one of:
a factor indicating that a prior call with the user was recently dropped,
a factor indicating a quantity of times the user has called within a particular time period, or
a factor indicating that the current communication is received from a location associated with a catastrophic event.
6. The method of claim 5 , wherein each factor, of the at least one factor, is associated with a respective weight; and
wherein the score for the current communication is based on each respective weight associated with each factor.
7. The method of claim 1 , further comprising:
ranking the current communication in the queue based on the score for the current communication and the other scores for the other communications included in the queue;
receiving second data identifying a second communication associated with a second user;
processing the second data and second historical communication data describing second prior communications associated with the second user, with the machine learning model, to determine a second score for the second communication; and
re-ranking the current communication in the queue based on the second score for the second communication.
8. A device, comprising:
one or more memories; and
one or more processors communicatively coupled to the one or more memories, configured to:
receive data identifying a current communication associated with a user;
process the data identifying the current communication and historical communication data describing prior communications associated with the user and received prior to the current communication, with a machine learning model, to determine a score for the current communication; and
place the current communication in a queue based on the score for the current communication and other scores for other communications included in the queue.
9. The device of claim 8 , wherein the one or more processors are further configured to:
receive other data identifying the other communications; and
process the other data and other historical communication data describing other prior communications associated with other users, with the machine learning model, to determine the other scores.
10. The device of claim 8 , wherein the one or more processors are further
configured to:
modify, in the queue and based on placing the current communication in the queue, at least one position of at least one other communication of the other communications.
11. The device of claim 8 , wherein the one or more processors, when
processing the data identifying the current communication and the historical communication data, are configured to:
determine, based on the historical communication data, a time period between the current communication and a historical communication identified by the historical communication data,
wherein the score for the current communication is based on the time period.
12. The device of claim 8 , wherein the machine learning model is associated with at least one factor used to determine the score for the current communication,
the at least one factor including at least one of:
a factor indicating that a prior call with the user was recently dropped,
a factor indicating a quantity of times the user has called within a particular time period, or
a factor indicating that the current communication is received from a location associated with a catastrophic event.
13. The device of claim 12 , wherein each factor, of the at least one factor, is associated with a respective weight; and
wherein the score for the current communication is based on each respective weight associated with each factor.
14. The device of claim 8 , wherein the one or more processors are further
configured to:
rank the current communication in the queue based on the score for the current communication
and the other scores for the other communications included in the queue;
receive second data identifying a second communication associated with a second user;
process the second data and second historical communication data describing second prior communications associated with the second user, with the machine learning model, to determine a second score for the second communication; and
re-rank the current communication in the queue based on the second score for the second communication.
15. A non-transitory computer-readable medium storing instructions, the
instructions comprising:
one or more instructions that, when executed by one or more processors, cause the one or more processors to:
receive data identifying a current communication associated with a user;
process the data identifying the current communication and historical communication data describing prior communications associated with the user and received prior to the current communication, with a machine learning model, to determine a score for the current communication; and
place the current communication in a queue based on the score for the current communication and other scores for other communications included in the queue.
16. The non-transitory computer-readable medium of claim 15 , wherein the one or more instructions, when executed by the one or more processors, further cause
the one or more processors to:
receive other data identifying the other communications; and
process the other data and other historical communication data describing other prior communications associated with other users, with the machine learning model, to determine the other scores.
17. The non-transitory computer-readable medium of claim 15 , wherein the
one or more instructions, when executed by the one or more processors, further cause the one or more processors to:
modify, in the queue and based on placing the current communication in the queue, at least one position of at least one other communication of the other communications.
18. The non-transitory computer-readable medium of claim 15 , wherein the
one or more instructions, that cause the one or more processors to process the data identifying the current communication and the historical communication data, cause the one or more processors to:
determine, based on the historical communication data, a time period between the current communication and a historical communication identified by the historical communication data,
wherein the score for the current communication is based on the time period.
19. The non-transitory computer-readable medium of claim 15 , wherein the
machine learning model is associated with at least one factor used to determine the score for the current communication,
the at least one factor including at least one of:
a factor indicating that a prior call with the user was recently dropped,
a factor indicating a quantity of times the user has called within a particular time period, or
a factor indicating that the current communication is received from a location associated with a catastrophic event.
20. The non-transitory computer-readable medium of claim 19 , wherein each factor, of the at least one factor, is associated with a respective weight; and
wherein the score for the current communication is based on each respective weight associated with each factor.