Methods and apparatus for campaign mapping for total audience measurement
Example methods and apparatus disclosed herein include campaign mapping for total audience measurement. An example apparatus includes a machine learning engine to predict sets of estimated duplication factors that represent duplicated media exposure across different possible combinations of media platforms for respective ones of a plurality of reference media campaigns, apply an input set of total exposure metrics associated with respective individual ones of the media platforms for a query media campaign to predict a first set of estimated duplication factors for the different possible combinations of media platforms for the query media campaign; identify a first one of the set of reference media campaigns to represent the query media campaign; and estimate a second set of estimated duplication factors for the query media campaign based on the set of estimated duplication factors for the first one of the set of reference media campaigns and the input set of total exposure metrics for the query media campaign.
1. An apparatus comprising:
memory; and
a machine learning engine to:
predict sets of estimated duplication factors that represent duplicated media exposure across different possible combinations of media platforms for respective ones of a plurality of reference media campaigns, the sets of estimated duplication factors corresponding to estimated measures of overlap of media exposure across the different possible combinations of the media platforms for the respective ones of the reference media campaigns, the machine learning engine trained to predict the sets of estimated duplication factors for the respective ones of the reference media campaigns from sets of total exposure metrics obtained for the respective ones of the reference media campaigns, the sets of total exposure metrics to represent media exposure associated with individual ones of the media platforms for the respective ones of the reference media campaigns, the machine learning engine trained based on the sets of total exposure metrics and sets of actual duplication factors obtained for the respective ones of the reference media campaigns, the sets of actual duplication factors different than the sets of estimated duplication factors, the sets of actual duplication factors corresponding to actual measures of overlap of media exposure across the different possible combinations of the media platforms for the respective ones of the reference media campaigns;
process an input set of total exposure metrics associated with respective individual ones of the media platforms for a query media campaign to predict a first set of estimated duplication factors that represent duplicated media exposure across the different possible combinations of media platforms for the query media campaign, the first set of estimated duplication factors corresponding to estimated measures of overlap of media exposure across the different possible combinations of the media platforms for the query media campaign;
identify a first reference media campaign of the reference media campaigns to represent the query media campaign based on comparisons of the first set of estimated duplication factors predicted for the query media campaign with respective ones of the sets of estimated duplication factors for the respective ones of the reference media campaigns; and
estimate a second set of estimated duplication factors for the query media campaign based on the set of estimated duplication factors for the first reference media campaign and the input set of total exposure metrics for the query media campaign, the second set of estimated duplication factors corresponding to estimated measures of overlap of media exposure across the different possible combinations of the media platforms for the query media campaign.
2. The apparatus of claim 1 , further including a duplication factor calculator to:
transform the sets of total exposure metrics and the sets of actual duplication factors to create a first reference media estimated duplication factor for a first combination of media platforms, the first reference media estimated duplication factor associated with a feature combination for the first combination of media platforms; and
transform the sets of total exposure metrics and the sets of actual duplication factors to create a second reference media estimated duplication factor for a second combination of media platforms, the second reference media estimated duplication factor associated with the feature combination.
3. The apparatus of claim 2 , wherein the feature combination represents at least one of demographics, a media campaign time step, a media platform reach, or a digital duplicated reach, the digital duplicated reach determined based on a combination of desktop reach, mobile reach, and digital reach.
4. The apparatus of claim 2 , further including a media platform engine to:
generate a first decision tree model based on the first reference media estimated duplication factor;
generate a second decision tree model based on the second reference media estimated duplication factor; and
train the first decision tree model and the second decision tree model using K-fold cross validation.
5. The apparatus of claim 2 , further including an embedder engine to transform the query media campaign based on the feature combination associated with the first reference media estimated duplication factor and the second reference media estimated duplication factor to create a query media estimated duplication factor.
6. The apparatus of claim 5 , further including a dimension engine to map the query media estimated duplication factor to the first reference media estimated duplication factor and the second reference media estimated duplication factor utilizing a KD tree, and identify the first reference media campaign based on a Euclidean distance to the query media campaign.
7. The apparatus of claim 6 , further including a max entropy engine to linearly interpolate the first reference media campaign to match a duration of the query media campaign, and combine the input set of total exposure metrics for the query media campaign and deduplication rates for the linearly interpolated reference media campaign to determine a deduplicated audience for the query media campaign.
8. A non-transitory computer readable medium comprising instructions that, when executed, cause a processor of a machine learning engine to at least:
train the machine learning engine to predict sets of estimated duplication factors that represent duplicated media exposure across different possible combinations of media platforms for respective ones of a plurality of reference media campaigns, the sets of estimated duplication factors corresponding to estimated measures of overlap of media exposure across the different possible combinations of the media platforms for the respective ones of the reference media campaigns, the machine learning engine trained to predict the sets of estimated duplication factors for the respective ones of the reference media campaigns from sets of total exposure metrics obtained for the respective ones of the reference media campaigns, the sets of total exposure metrics to represent media exposure associated with individual ones of the media platforms for the respective ones of the reference media campaigns, the machine learning engine trained based on the sets of total exposure metrics and sets of actual duplication factors obtained for the respective ones of the reference media campaigns, the sets of actual duplication factors different than the sets of estimated duplication factors, the sets of actual duplication factors corresponding to actual measures of overlap of media exposure across the different possible combinations of the media platforms for the respective ones of the reference media campaigns;
process an input set of total exposure metrics associated with respective individual ones of the media platforms for a query media campaign to predict a first set of estimated duplication factors that represent duplicated media exposure across the different possible combinations of media platforms for the query media campaign, the first set of estimated duplication factors corresponding to estimated measures of overlap of media exposure across the different possible combinations of the media platforms for the query media campaign;
identify a first reference media campaign of the reference media campaigns to represent the query media campaign based on comparisons of the first set of estimated duplication factors predicted for the query media campaign with respective ones of the sets of estimated duplication factors for the respective ones of the reference media campaigns; and
estimate a second set of estimated duplication factors for the query media campaign based on the set of estimated duplication factors for the first reference media campaign and the input set of total exposure metrics for the query media campaign, the second set of estimated duplication factors corresponding to estimated measures of overlap of media exposure across the different possible combinations of the media platforms for the query media campaign.
9. The non-transitory computer readable medium of claim 8 , wherein the instructions further cause the processor of the machine learning engine to:
transform the sets of total exposure metrics and the sets of actual duplication factors to create a first reference media estimated duplication factor for a first combination of media platforms, the first reference media estimated duplication factor associated with a feature combination for the first combination of media platforms; and
transform the sets of total exposure metrics and the sets of actual duplication factors to create a second reference media estimated duplication factor for a second combination of media platforms, the second reference media estimated duplication factor associated with the feature combination.
10. The non-transitory computer readable medium of claim 9 , wherein the feature combination represents at least one of demographics, a media campaign time step, a media platform reach, or a digital duplicated reach, the digital duplicated reach determined based on a combination of desktop reach, mobile reach, and digital reach.
11. The non-transitory computer readable medium of claim 9 , wherein the instructions further cause the processor of the machine learning engine to:
generate a first decision tree model based on the first reference media estimated duplication factor;
generate a second decision tree model based on the second reference media estimated duplication factor; and
train the first decision tree model and the second decision tree model using K-fold cross validation.
12. The non-transitory computer readable medium of claim 9 , wherein the instructions further cause the processor of the machine learning engine to transform the query media campaign based on the feature combination associated with the first reference media estimated duplication factor and the second reference media estimated duplication factor to create a query media estimated duplication factor.
13. The non-transitory computer readable medium of claim 12 , wherein the instructions further cause the processor of the machine learning engine to map the query media estimated duplication factor to the first reference media estimated duplication factor and the second reference media estimated duplication factor utilizing a KD tree, and identify the first reference media campaign based on a Euclidean distance to the query media campaign.
14. The non-transitory computer readable medium of claim 13 , wherein the instructions further cause the processor of the machine learning engine to linearly interpolate the first reference media campaign to match a duration of the query media campaign, and combine the input set of total exposure metrics for the query media campaign and deduplication rates for the linearly interpolated reference media campaign to determine a deduplicated audience for the query media campaign.
15. A method comprising:
training, by executing an instruction with a processor, a machine learning engine to predict sets of estimated duplication factors that represent duplicated media exposure across different possible combinations of media platforms for respective ones of a plurality of reference media campaigns, the sets of estimated duplication factors corresponding to estimated measures of overlap of media exposure across the different possible combinations of the media platforms for the respective ones of the reference media campaigns, the machine learning engine trained to predict the sets of estimated duplication factors for the respective ones of the reference media campaigns from sets of total exposure metrics obtained for the respective ones of the reference media campaigns, the sets of total exposure metrics to represent media exposure associated with individual ones of the media platforms for the respective ones of the reference media campaigns, the machine learning engine trained based on the sets of total exposure metrics and sets of actual duplication factors obtained for the respective ones of the reference media campaigns, the sets of actual duplication factors different than the sets of estimated duplication factors, the sets of actual duplication factors corresponding to actual measures of overlap of media exposure across the different possible combinations of the media platforms for the respective ones of the reference media campaigns;
processing, with the machine learning engine, an input set of total exposure metrics associated with respective individual ones of the media platforms for a query media campaign to predict a first set of estimated duplication factors that represent duplicated media exposure across the different possible combinations of media platforms for the query media campaign, the first set of estimated duplication factors corresponding to estimated measures of overlap of media exposure across the different possible combinations of the media platforms for the query media campaign;
identifying, with the machine learning engine, a first reference media campaign of the reference media campaigns to represent the query media campaign based on comparisons of the first set of estimated duplication factors predicted for the query media campaign with respective ones of the sets of estimated duplication factors for the respective ones of the reference media campaigns; and
estimating, with the machine learning engine, a second set of estimated duplication factors for the query media campaign based on the set of estimated duplication factors for the first reference media campaign and the input set of total exposure metrics for the query media campaign, the second set of estimated duplication factors corresponding to estimated measures of overlap of media exposure across the different possible combinations of the media platforms for the query media campaign.
16. The method of claim 15 , wherein the training of the machine learning engine includes:
transforming the sets of total exposure metrics and the sets of actual duplication factors to create a first reference media estimated duplication factor for a first combination of media platforms, the first reference media estimated duplication factor associated with a feature combination for the first combination of media platforms; and
transforming the sets of total exposure metrics and the sets of actual duplication factors to create a second reference media estimated duplication factor for a second combination of media platforms, the second reference media estimated duplication factor associated with the feature combination.
17. The method of claim 16 , further including:
generating a first decision tree model based on the first reference media estimated duplication factor;
generating a second decision tree model based on the second reference media estimated duplication factor; and
training the first decision tree model and the second decision tree model using K-fold cross validation.
18. The method of claim 16 , further including:
transforming the query media campaign based on the feature combination associated with the first reference media estimated duplication factor and the second reference media estimated duplication factor to create a query media estimated duplication factor.
19. The method of claim 18 , further including mapping the query media estimated duplication factor to the first reference media estimated duplication factor and the second reference media estimated duplication factor utilizing a KD tree, and identify the first reference media campaign based on a Euclidean distance to the query media campaign.
20. The method of claim 19 , wherein the estimating of the second set of estimated duplication factors for the query media campaign includes linearly interpolating the first reference media campaign to match a duration of the query media campaign, and further including combining the input set of total exposure metrics for the query media campaign and deduplication rates for the linearly interpolated reference media campaign to determine a deduplicated audience for the query media campaign.
21. The method of claim 16 , wherein the feature combination represents at least one of demographics, a media campaign time step, a media platform reach, or a digital duplicated reach, the digital duplicated reach determined based on a combination of desktop reach, mobile reach, and digital reach.
22. The apparatus of claim 1 , wherein the reference media campaigns correspond to campaigns for which the overlap of the media exposure across the different possible combinations of the media platforms is known, and the query media campaign corresponds to a campaign for which the overlap of the media exposure across the different possible combinations of the media platforms is unknown.