System and method for reducing subscriber churn through complimentary data allocations

1 . A method of allocating complimentary data to subscribers of a wireless network, comprising:

identifying, from a plurality of subscribers of the wireless network, one or more subscribers that each have used at least a first predetermined percentage of their respective data quotas for a billing cycle, wherein the first predetermined percentage is less than their respective data quota;

predicting, using a machine learning model, that at least one subscriber of the one or more subscribers is to exceed the respective data quota by the end of the billing cycle;

determining a size of complementary data to be allocated to each of the at least one subscriber such that at least a second percentage of the at least one subscriber is not to exceed the respective data quota; and

allocating the complimentary data with the determined size to the at least one subscriber through one of a plurality of network functions in the wireless network.

2 . The method of claim 1 , wherein each subscriber of the plurality of subscribers is on an unlimited data plan with the same data quota for the billing cycle, and wherein data throttling occurs after the respective data quota is reached.

3 . The method of claim 1 , wherein the machine learning model is a deep learning model.

4 . The method of claim 1 , wherein the wireless network is a 5th generation (5G) network.

5 . The method of claim 1 , wherein the determining of the size of the complementary data to be allocated to each of the at least one subscriber is performed using quantile regression.

6 . The method of claim 1 , wherein the network function that allocates the complimentary data with the determined size is a session management function (SMF) of the wireless network.

7 . The method of claim 1 , wherein the first predetermined percentage is smaller than the second predetermined percentage.

8 . The method of claim 1 , wherein the identifying, predicting, determining, and allocating are repeated at least once by the end of the billing cycle.

9 . A system, comprising:

one or more processors; and

one or more memories that are coupled to the one or more processors and storing program instructions for allocating complimentary data to subscribers of a wireless network, wherein the program instructions, when executed by the one or more processors, cause the system to perform operations comprising:

identifying, from a plurality of subscribers of the wireless network, one or more subscribers that each have used at least a first predetermined percentage of their respective data quotas for a billing cycle, wherein the first predetermined percentage is less than their respective data quota;

predicting, using a machine learning model, that at least one subscriber of the one or more subscribers is to exceed the respective data quota by the end of the billing cycle;

determining a size of complementary data to be allocated to each of the at least one subscriber such that at least a second percentage of the at least one subscriber is not to exceed the respective data quota; and

allocating the complimentary data with the determined size to the at least one subscriber through one of a plurality of network functions in the wireless network.

10 . The system of claim 9 , wherein each subscriber of the plurality of subscribers is on an unlimited data plan with the same data quota for the billing cycle, and wherein data throttling occurs after the respective data quota is reached.

11 . The system of claim 9 , wherein the machine learning model is a deep learning model.

12 . The system of claim 9 , wherein the wireless network is a 5th generation (5G) network.

13 . The system of claim 9 , wherein the determining of the size of the complementary data to be allocated to each of the at least one subscriber is performed using quantile regression.

14 . The system of claim 9 , wherein the network function that allocates the complimentary data with the determined size is a session management function (SMF) of the wireless network.

15 . The system of claim 9 , wherein the first predetermined percentage is smaller than the second predetermined percentage.

16 . The system of claim 9 , wherein the identifying, predicting, determining, and allocating are repeated at least once by the end of the billing cycle.

17 . A non-transitory computer readable storage medium storing program instructions for allocating complimentary data to subscribers of a wireless network, wherein the program instructions, when executed by one or more processors, cause the one or more processors to perform operations comprising:

identifying, from a plurality of subscribers of the wireless network, one or more subscribers that each have used at least a first predetermined percentage of their respective data quotas for a billing cycle, wherein the first predetermined percentage is less than their respective data quota;

predicting, using a machine learning model, that at least one subscriber of the one or more subscribers is to exceed the respective data quota by the end of the billing cycle;

determining a size of complementary data to be allocated to each of the at least one subscriber such that at least a second percentage of the at least one subscriber is not to exceed the respective data quota; and

allocating the complimentary data with the determined size to the at least one subscriber through one of a plurality of network functions in the wireless network.

18 . The non-transitory computer readable storage medium of claim 17 , wherein each subscriber of the plurality of subscribers is on an unlimited data plan with the same data quota for the billing cycle, and wherein data throttling occurs after the respective data quota is reached.

19 . The non-transitory computer readable storage medium of claim 17 , wherein the machine learning model is a deep learning model.

20 . The non-transitory computer readable storage medium of claim 17 , wherein the wireless network is a 5th generation (5G) network.