Generation and utilization of channel allocation models for resource allocation recommendations

1 . A computer-implemented method comprising:

obtaining data associated with a plurality of media channels;

inputting the data into a machine learning model;

estimating, by the machine learning model, a structure of a causal graph comprising a Bayesian belief network (BBN) by:

performing structure learning to generate the causal graph comprising a plurality of nodes and a plurality of edges by calculating, by a random forest, a variable importance for the plurality of nodes to assist in refining the structure of the causal graph to favor an inclusion of a first set of relationships between a first set of the plurality of nodes and disfavor an inclusion of a second set of relationships between a second set of the plurality of nodes;

applying a second machine learning model to the causal graph to estimate one or more parameters associated with the causal graph;

determining an allocation of resources to the plurality of media channels based on the causal graph by:

generating a plurality of simulations for a plurality of budget allocation scenarios based on the causal graph comprising the BBN;

determining, for each respective simulation of the plurality of simulations, a parameter; and

determining, based on the parameter for each respective simulation of the plurality of simulations compared to a target parameter, an optimal budget allocation; and

allocating the resources to the plurality of media channels based on the determined resource allocation, wherein the resources comprise at least one of: network resources or computing resources.

2 . The method of claim 1 , wherein the second machine learning model comprises a kernel-based machine learning model.

3 . The method of claim 1 , wherein the structure learning comprises:

applying a machine learning model for feature selection; and

performing bootstrap aggregation for updating the causal graph.

4 . The method of claim 1 , wherein estimating the structure of the causal graph comprises: performing parameter learning to determine a (i) magnitude and (ii) direction for each respective edge of the plurality of edges.

5 . The method of claim 4 , wherein performing the parameter learning comprises:

performing parameter regularization.

6 . The method of claim 1 , wherein the causal graph comprises a directed acyclic graph representing conditional probabilities between the nodes.

7 . The method of claim 1 , comprising performing a validation method by:

transmitting data comprising instructions that cause a graphical representation of the causal graph to display via a graphical user interface of a device associated with a user; and

obtaining data indicative of user input of acceptance of the graphical representation of the causal graph.

8 . The method of claim 1 , wherein determining the allocation of resources to the plurality of media channels based on the causal graph comprises:

determining a relationship between a first node, a second node, and a third node; and

based on the relationship between the first node, the second node, and the third node, determining an optimal budget allocation for a first media channel associated with the first node, a second media channel associated with the second node, and a third media channel associated with the third node.

9 . The method of claim 8 , wherein determining the optimal budget allocation for the first media channel, the second media channel, and the third media channel comprises:

simulating data indicative of a plurality of budget allocation scenarios, wherein each budget allocation scenario is indicative of an amount of budget allocated to each of the first media channel, the second media channel, and the third media channel;

obtaining data indicative of user input of a target parameter;

for each respective budget allocation scenario of the plurality of budget allocation scenarios, determining a respective parameter for a respective budget allocation scenario of the plurality of budget allocation scenarios;

comparing each respective parameter and the target parameter;

determining that a first respective parameter associated with a first budget allocation scenario is within a threshold of the target parameter; and

in response to determining that the first respective parameter is within the threshold of the target parameter, selecting the first budget allocation scenario as an optimal budget allocation scenario.

10 . The method of claim 9 , wherein the target parameter and each respective parameter associated with each respective budget allocation scenario are indicative of a percent reduction in cost per sale.

11 . The method of claim 10 , wherein the cost per sale is determined by calculating a media spend by the respective media channel divided by unit sales driven by the respective media channel.

12 . The method of claim 9 , wherein the target parameter and each respective parameter associated with each respective budget allocation scenario are indicative of an attribution associated with an effectiveness of the respective media channel and promotion usage on target actions.

13 . The method of claim 12 , wherein the attribution for the respective media channel is determined by calculating a unit sale driven by the respective media channel divided by overall sales.

14 . The method of claim 9 , wherein the target parameter and each respective parameter associated with each respective budget allocation scenario are an overall total media spend budget.

15 . The method of claim 1 , comprising:

generating data indicative of a graphical representation of the causal graph; and

transmitting data comprising instructions that, when executed, cause the graphical representation of the causal graph to render via a graphical user interface of a user device.

16 . The method of claim 1 , comprising:

generating data indicative of a graphical representation of the determined allocation of resources to the plurality of media channels; and

transmitting data comprising instructions that, when executed, cause the graphical representation of the allocation of resources to the plurality of media channels to render via a graphical user interface of a user device.

17 . The method of claim 1 , where the data associated with the plurality of media channels comprises outcome data, predictive variables, and control variables.

18 . A computing system, comprising:

one or more processors; and

one or more computer-readable media storing instructions that are executable to cause the one or more processors to perform operations, the operations comprising:

obtaining data associated with a plurality of media channels;

inputting the data into a machine learning model;

estimating, by the machine learning model, a structure of a causal graph comprising a Bayesian belief network (BBN) by:

performing structure learning to generate the causal graph comprising a plurality of nodes and a plurality of edges by calculating, by a random forest, a variable importance for the plurality of nodes to assist in refining the structure of the causal graph to favor an inclusion of a first set of relationships between a first set of the plurality of nodes and disfavor an inclusion of a second set of relationships between a second set of the plurality of nodes;

applying a second machine learning model to the causal graph to estimate one or more parameters associated with the causal graph;

determining an allocation of resources to the plurality of media channels based on the causal graph by:

generating a plurality of simulations for a plurality of budget allocation scenarios based on the causal graph comprising the BBN;

determining, for each respective simulation of the plurality of simulations, a parameter; and

determining, based on the parameter for each respective simulation of the plurality of simulations compared to a target parameter, an optimal budget allocation; and

allocating the resources to the plurality of media channels based on the determined resource allocation, wherein the resources comprise at least one of: network resources or computing resources.

19 . One or more non-transitory computer readable media storing instructions that are executable by one or more processors to perform operations comprising:

obtaining data associated with a plurality of media channels;

inputting the data into a machine learning model;

estimating, by the machine learning model, a structure of a causal graph comprising a Bayesian belief network (BBN) by:

performing structure learning to generate the causal graph comprising a plurality of nodes and a plurality of edges by calculating, by a random forest, a variable importance for the plurality of nodes to assist in refining the structure of the causal graph to favor an inclusion of a first set of relationships between a first set of the plurality of nodes and disfavor an inclusion of a second set of relationships between a second set of the plurality of nodes;

applying a second machine learning model to the causal graph to estimate one or more parameters associated with the causal graph;

determining an allocation of resources to the plurality of media channels based on the causal graph by:

generating a plurality of simulations for a plurality of budget allocation scenarios based on the causal graph comprising the BBN;

determining, for each respective simulation of the plurality of simulations, a parameter; and

determining, based on the parameter for each respective simulation of the plurality of simulations compared to a target parameter, an optimal budget allocation; and

allocating the resources to the plurality of media channels based on the determined resource allocation, wherein the resources comprise at least one of: network resources or computing resources.