Labeling apparatus

1. A method comprising:

receiving a data set that includes demographic data and population data;

training, based on the data set, a neural network to establish a relationship between different physical layouts of messages and responses to the different physical layouts of the messages, the training of the neural network comprising:

identifying, based on the population data, a first population set that responds to messages after being presented with a first quantity of identical messages;

identifying, based on the population data, a second population set that responds to the messages after being presented with a second quantity of identical messages; and

optimizing the neural network to determine a specified quantity of messages to deliver to a given user prior to defaulting to communicating an alternate message to the given user;

applying the trained neural network to a user profile to predict a physical layout of a message;

generating instructions for an electronic device based on the predicted physical layout of the message, the instructions comprising the message; and

transmitting the instructions to the electronic device to create a physical label having a layout corresponding to the predicted physical layout of the message.

2. The method of claim 1 , comprising:

minimizing a loss function to update parameters of the neural network.

3. The method of claim 1 , wherein the different physical layouts of messages comprise messages having different sizes and different shapes.

4. The method of claim 1 , wherein a refill message is prioritized above an adherence message in response to determining that the given user has no remaining refills.

5. The method of claim 1 , wherein the neural network is trained through a linear regression model.

6. A system comprising:

a memory; and

one or more processors coupled to the memory and configured to execute instructions stored in the memory to perform operations comprising:

receiving a data set that includes demographic data and population data;

training, based on the data set, a neural network to establish a relationship between different physical layouts of messages and responses to the different physical layouts of the messages, the training of the neural network comprising:

identifying, based on the population data, a first population set that responds to messages after being presented with a first quantity of identical messages;

identifying, based on the population data, a second population set that responds to the messages after being presented with a second quantity of identical messages; and

optimizing the neural network to determine a specified quantity of messages to deliver to a given user prior to defaulting to communicating an alternate message to the given user;

applying the trained neural network to a user profile to predict a physical layout of a message;

generating instructions for an electronic device based on the predicted physical layout of the message, the instructions comprising the message; and

transmitting the instructions to the electronic device to create a physical label having a layout corresponding to the predicted physical layout of the message.

7. The system of claim 6 , the operations comprising:

minimizing a loss function to update parameters of the neural network.

8. The system of claim 6 , wherein the different physical layouts of messages comprise messages having different sizes and different shapes.

9. The system of claim 6 , wherein a refill message is prioritized above an adherence message in response to determining that the given user has no remaining refills.

10. The system of claim 6 , wherein the neural network is trained through a linear regression model.

11. A method comprising:

receiving a data set that includes demographic data and population data;

training, based on the data set, a neural network to establish a relationship between different physical layouts of messages and responses to the different physical layouts of the messages, the different physical layouts of messages comprise messages having different sizes and different shapes, the training of the neural network comprising:

identifying, based on the population data, a first population set that responds to messages after being presented with a first quantity of identical messages;

identifying, based on the population data, a second population set that responds to the messages after being presented with a second quantity of identical messages;

minimizing a loss function to update parameters of the neural network based on the first and second population sets; and

optimizing the neural network to determine a specified quantity of messages to deliver to a given user prior to defaulting to communicating an alternate message to the given user;

applying the trained neural network to a user profile to predict a physical layout of a message;

generating instructions for an electronic device based on the predicted physical layout of the message, the instructions comprising the message; and

transmitting the instructions to the electronic device to create a physical label having a layout corresponding to the predicted physical layout of the message, wherein a refill message is prioritized above an adherence message in response to determining that the given user has no remaining refills.

12. The method of claim 11 , comprising:

minimizing a loss function to update parameters of the neural network.

13. The method of claim 11 , wherein the different physical layouts of messages comprise messages having different sizes and different shapes.

14. The method of claim 11 , wherein a refill message is prioritized above an adherence message in response to determining that the given user has no remaining refills.

15. The method of claim 11 , wherein the neural network is trained through a linear regression model.