METHODS AND APPARATUS TO GENERATE AUDIENCE METRICS USING THIRD-PARTY PRIVACY-PROTECTED CLOUD ENVIRONMENTS

1 . An apparatus comprising:

at least one memory;

instructions; and

processor circuitry to execute the instructions to at least:

generate an individualization model based on truth data indicating first true users exposed to media via first panelist client devices;

produce user probabilities for second panelist client devices based on the individualization model, the user probabilities indicating likelihoods of second true users being exposed to media via the second panelist client devices;

select a user probability from the user probabilities based on an impression, the impression associated with at least one feature and a selected device from the second panelist client devices, the user probability associated with the at least one feature and the selected device, the user probability indicating a likelihood of a set of second true users being exposed to media corresponding to the impression via the selected device; and

assign the impression to the set of second true users.

2 . The apparatus of claim 1 , wherein the user probabilities are first user probabilities, wherein the user probability is a first user probability, wherein the impression is a first impression, wherein the selected device is a first selected device,

wherein the processor circuitry is to further:

produce second user probabilities for the first panelist client devices based on the truth data;

select a second user probability from the second user probabilities based on a second impression associated with at least one feature and a second selected device from the first panelist client devices, the second user probability associated with the at least one feature and the second selected device, the second user probability indicating a likelihood of one or more first true users being exposed to media corresponding to the impression via the second selected device; and

assign the second impression to the one or more first true users.

3 . The apparatus of claim 1 , wherein the processor circuitry is to:

generate the individualization model is based on training and validating the individualization model; and

train and validate the individualization model based on the truth data.

4 . The apparatus of claim 1 , wherein the processor circuitry is to:

produce the user probabilities is further based on predicting demographic probabilities; and

predict demographic probabilities associated with the second panelist client devices based on the individualization model, the demographic probabilities indicating likelihoods of demographics corresponding to the second true users being exposed to the media via the second panelist client devices.

5 . The apparatus of claim 4 , wherein the demographics include at least one of age, gender, race, income, home location, or occupation.

6 . The apparatus of claim 1 , wherein at least one feature includes at least one of type of content, a time, or a type of device.

7 . The apparatus of claim 1 , wherein the truth data is from survey responses indicating at least one of a type of content, a time of day, or a type of device, associated with the first true users.

8 . The apparatus of claim 1 , wherein the processor circuitry is to produce the user probabilities is further based on known information including at least one of primary users, type of devices, or demographic composition corresponding to the second panelist client devices.

9 . The apparatus of claim 1 , wherein the set of second true users include more than one second true users, wherein processor circuitry is to assign fractional impressions to the set of second true users to assign the impression to the set of second true users.

10 . The apparatus of claim 1 , wherein processor circuitry is to store the individualization model to at least one memory.

11 . The apparatus of claim 1 , wherein the processor circuitry is to store the user probabilities to at least one memory.

12 . A non-transitory computer readable medium comprising instructions which, when executed, cause at least one processor to:

generate an individualization model based on truth data indicating first true users exposed to media via first panelist client devices;

produce user probabilities for second panelist client devices based on the individualization model, the user probabilities indicating likelihoods of second true users being exposed to media via the second panelist client devices;

select a user probability from the user probabilities based on an impression, the impression associated with at least one feature and a selected device from the second panelist client devices, the user probability associated with the at least one feature and the selected device, the user probability indicating a likelihood of a set of second true users being exposed to media corresponding to the impression via the selected device; and

assign the impression to the set of second true users.

13 . The computer readable medium of claim 12 , wherein the user probabilities are first user probabilities, wherein the user probability is a first user probability, wherein the impression is a first impression, wherein the selected device is a first selected device, wherein the instructions cause the at least one processor to:

produce second user probabilities for the first panelist client devices based on the truth data;

select a second user probability from the second user probabilities based on a second impression associated with at least one feature and a second selected device from the first panelist client devices, the second user probability associated with the at least one feature and the second selected device, the second user probability indicating a likelihood of one or more first true users being exposed to media corresponding to the impression via the second selected device; and

assign the second impression to the one or more first true users.

14 . The computer readable medium of claim 12 , wherein the instructions cause the at least one processor to:

generate the individualization model is based on training and validating the individualization model; and

train and validate the individualization model based on the truth data.

15 . The computer readable medium of claim 12 , wherein the instructions cause the at least one processor to:

produce the user probabilities is further based on predicting demographic probabilities; and

predict demographic probabilities associated with the second panelist client devices based on the individualization model, the demographic probabilities indicating likelihoods of demographics corresponding to the second true users being exposed to the media via the second panelist client devices.

16 . The computer readable medium of claim 15 , wherein the demographics include at least one of age, gender, race, income, home location, or occupation.

17 . The computer readable medium of claim 12 , wherein at least one feature includes at least one of type of content, a time, or a type of device.

18 . The computer readable medium of claim 12 , wherein the truth data is from survey responses indicating at least one of a type of content, a time of day, or a type of device, associated with the first true users.

19 . The computer readable medium of claim 12 , wherein the instructions cause the at least one processor to produce the user probabilities is further based on known information including at least one of primary users, type of devices, or demographic composition corresponding to the second panelist client devices.

20 . The computer readable medium of claim 12 , wherein the set of second true users include more than one second true users, wherein the instructions cause the at least one processor to assign fractional impressions to the set of second true users to assign the impression to the set of second true users.