Methods and apparatus to determine demographic classifications for census level impression counts and unique audience sizes

1. An audience measurement computing system comprising:

a database;

a network interface that communicatively couples the database to a processor; and

the processor configured to, upon executing machine readable instructions, cause the audience measurement computing system to perform operations including one or more of:

generating a reformatted media tag corresponding to an impression request;

executing a machine learning model based on the reformatted media tag to generate outputs, the outputs including at least a value representative of a probability of an occurrence of a demographic classification; and

assigning an identification of ones of audience members in a group to the demographic classification based at least on the outputs.

2. The audience measurement computing system of claim 1 , wherein the operations further include:

matching identification data in a media tag to personal information in the database to thereby generate the reformatted media tag; and

training the machine learning model on the reformatted media tag.

3. The audience measurement computing system of claim 2 , wherein training the machine learning model includes:

comparing the probability of the occurrence of the demographic classification to a known population of audience members in the group; and

adjusting at least one parameter of the machine learning model based on the comparison of the probability of the occurrence of the demographic classification to the known population of audience members in the group.

4. The audience measurement computing system of claim 3 , wherein training the machine learning model further includes masking the outputs when the probability of the occurrence of the demographic classification satisfies an inconsistency threshold level against the known population of audience members in the group.

5. The audience measurement computing system of claim 1 , wherein generating the reformatted media tag further includes:

generating the reformatted media tag corresponding to the impression request based on input parameters of the machine learning model; and

appending ones of features to the reformatted media tag.

6. The audience measurement computing system of claim 1 , wherein the value is a first value, the demographic classification is a first demographic classification, the outputs include a plurality of values, the plurality of values including the first value, ones of the plurality of values are representative of ones of probabilities of occurrences of a plurality of demographic classifications, and the plurality of demographic classifications include the first demographic classification.

7. The audience measurement computing system of claim 6 , wherein the operations further include incrementing a count of unique audience members in the group representing ones of the plurality of demographic classifications based on the ones of probabilities.

8. The audience measurement computing system of claim 7 , wherein the impression request is a first impression request of a plurality of impression requests, the group is a first group of a plurality of groups, the count of unique audience members is a first count of unique audience members, and wherein the operations further include:

aggregating ones of the plurality of impression requests and ones of counts of unique audience members representative of ones of groups of the plurality of groups; and

generating a report of a distribution of demographic classifications across the ones of groups.

9. A non-transitory machine readable storage medium comprising instructions that, when executed, cause a processor to be configured to at least:

generate a reformatted media tag corresponding to an impression request;

execute a machine learning model based on the reformatted media tag to generate outputs, the outputs including at least a value representative of a probability of an occurrence of a demographic classification; and

assign an identification of ones of audience members in a group to the demographic classification based at least on the outputs.

10. The non-transitory machine readable storage medium of claim 9 , wherein the instructions further cause the processor to be configured to at least:

match identification data in a media tag to personal information in a database to thereby generate the reformatted media tag; and

train the machine learning model based on the reformatted media tag.

11. The non-transitory machine readable storage medium of claim 10 , wherein the instructions further cause the processor to be configured to at least:

compare the probability of the occurrence of the demographic classification to a known population of audience members in the group; and

adjust at least one parameter of the machine learning model based on the comparison of the probability of the occurrence of the demographic classification to a known population of audience members in the group.

12. The non-transitory machine readable storage medium of claim 11 , wherein the instructions further cause the processor to be configured to at least mask the outputs when the probability of the occurrence of the demographic classification satisfies an inconsistency threshold level against the known population of audience members in the group.

13. The non-transitory machine readable storage medium of claim 9 , wherein the instructions further cause the processor to be configured to at least:

generate the reformatted media tag corresponding to the impression request based on input parameters of the machine learning model; and

append ones of features to the reformatted media tag.

14. The non-transitory machine readable storage medium of claim 9 , wherein the value is a first value and the demographic classification is a first demographic classification, the outputs include a plurality of values, the plurality of values including the first value, ones of the plurality of values are representative of ones of probabilities of occurrences of a plurality of demographic classifications, and the plurality of demographic classifications include the first demographic classification.

15. The non-transitory machine readable storage medium of claim 14 , wherein the instructions further cause the processor to be configured to at least:

increment a count of unique audience members in the group representing ones of the plurality of demographic classifications based on the ones of probabilities.

16. The non-transitory machine readable storage medium of claim 15 , wherein the impression request is a first impression request of a plurality of impression requests, the group is a first group of a plurality of groups, the count of unique audience members is a first count of unique audience members, wherein the instructions further cause the processor to be configured to at least:

aggregate ones of the plurality of impression requests and ones of counts of unique audience members representative of ones of groups of the plurality of groups; and

generate a report of a distribution of demographic classifications across the ones of groups.

17. A method, comprising:

generating a reformatted media tag corresponding to an impression request;

executing a machine learning model based on the reformatted media tag to generate outputs, the outputs including at least a value representative of a probability of an occurrence of a demographic classification; and

assigning an identification of ones of audience members in a group to the demographic classification based at least on the outputs.

18. The method of claim 17 , further including:

matching identification data in a media tag to personal information in a database to thereby generate the reformatted media tag; and

training the machine learning model based on the reformatted media tag.

19. The method claim 18 , further including:

comparing the probability of the occurrence of the demographic classification to a known population of audience members in the group; and

adjusting at least one parameter of the machine learning model based on the comparison of the probability of the occurrence of the demographic classification to a known population of audience members in the group.

20. The method of claim 19 , further including:

masking the outputs when the probability of the occurrence of the demographic classification satisfies an inconsistency threshold level against the known population of audience members in the group.

21. The method of claim 17 , further including:

generating the reformatted media tag corresponding to the impression request based on input parameters of the machine learning model; and

appending ones of features to the reformatted media tag.

22. The method of claim 17 , wherein the value is a first value and the demographic classification is a first demographic classification, the outputs including a plurality of values, the plurality of values including the first value, ones of the plurality of values are representative of ones of probabilities of occurrences of a plurality of demographic classifications, and the plurality of demographic classifications include the first demographic classification.

23. The method claim 22 , further including:

incrementing a count of unique audience members in the group representing ones of the plurality of demographic classifications based on the ones of probabilities.

24. The method of claim 23 , wherein the impression request is a first impression request of a plurality of impression requests, the group is a first group of a plurality of groups, the count of unique audience members is a first count of unique audience members, and wherein the method further includes:

aggregating ones of the plurality of impression requests and ones of counts of unique audience members representative of ones of groups of the plurality of groups; and

generating a report of a distribution of demographic classifications across the ones of groups.