Methods and apparatus to generate audience metrics for connected television
Methods, apparatus, systems, and articles of manufacture are disclosed to generate audience metrics for connected television. An example system includes at least one memory, programmable circuitry, and instructions to cause the programmable circuitry to obtain media access data corresponding to connected television media and a user identifier corresponding to a media access device, generate, using a machine learning model, probability values for corresponding audience demographics in a household composition corresponding to the user identifier, the probability values indicative of likelihoods that corresponding ones of the audience demographics are accessing the connected television media, determine a person-level characteristic based on the probability values of the audience demographics, the person-level characteristic corresponding to an audience member of the connected television media, and assign the media access data to the person-level characteristic.
1. An audience measurement computing system comprising:
a processor; and
at least one memory storing instructions that, when executed by the processor, cause the audience measurement computing system to perform operations comprising:
obtaining media access data corresponding to connected television media and a user identifier corresponding to a media access device, wherein the user identifier comprises a publisher-assigned user identifier;
obtaining third-party data corresponding to the publisher-assigned user identifier from one or more database proprietors, the third-party data comprising information associated with a household composition corresponding to the publisher-assigned user identifier,
providing, as an input to a trained machine learning model, the third-party data corresponding to the publisher-assigned user identifier obtained from the one or more database proprietors, the media access data, and the publisher-assigned user identifier;
generating, using the trained machine learning model and based on the input, probability values for corresponding audience demographics in the household composition corresponding to the publisher-assigned user identifier, the probability values indicative of likelihoods that corresponding ones of the audience demographics are accessing the connected television media;
determining a person-level characteristic based on the probability values of the audience demographics, the person-level characteristic corresponding to an audience member of the connected television media;
assigning media features extracted from the media access data to the person-level characteristic;
mapping the publisher-assigned user identifier to the person-level characteristic corresponding to the audience member of the connected television media; and
transmitting the publisher-assigned user identifier and the person-level characteristic to a publisher of the connected television media.
2. The audience measurement computing system of claim 1 , wherein the operations further comprise:
identifying the household composition based on the user identifier, the household composition indicative of one or more demographic categories that use the media access device.
3. The audience measurement computing system of claim 1 , wherein the person-level characteristic includes at least one of an age or a gender.
4. The audience measurement computing system of claim 1 , wherein the operations further comprise:
analyzing historical media access data to determine the household composition associated with the media access device, the historical media access data generated during previous media access sessions presented by the media access device.
5. The audience measurement computing system of claim 4 , wherein the media access data is substantially real-time media access data, and wherein the operations further comprise:
generating a mapping of the household composition with the user identifier, the mapping to subsequently generate the probability values based on the substantially real-time media access data and the user identifier.
6. The audience measurement computing system of claim 1 , wherein the operations further comprise:
generating a report including the audience member of the connected television media, the report to provide substantially real-time feedback of the audience member to an advertiser.
7. The audience measurement computing system of claim 1 , wherein the operations further comprise:
obtaining an application programming interface call from the publisher of the connected television media, the application programming interface call indicative of second instructions to identify the audience member of the connected television media in substantially real-time.
8. The audience measurement computing system of claim 1 , wherein the operations further comprise:
determining the person-level characteristic based on the person-level characteristic corresponding to a largest one of the probability values.
9. The audience measurement computing system of claim 1 , wherein the operations further comprise:
generating, using a neural network, scores for demographic categories based on historical media access data; and
comparing the scores to a threshold to determine the household composition.
10. The system of claim 1 , wherein the operations further comprise, before assigning the media features to the person-level characteristic:
processing the media access data by using filters to extract the media features from the media access data.
11. A non-transitory machine readable storage medium comprising instructions that, when executed, cause a processor to perform operations comprising:
obtaining media access data corresponding to connected television media and a user identifier corresponding to a media access device, wherein the user identifier comprises a publisher-assigned user identifier;
obtaining third-party data corresponding to the publisher-assigned user identifier from one or more database proprietors, the third-party data comprising information associated with a household composition corresponding to the publisher-assigned user identifier, providing, as an input to a trained machine learning model, the third-party data corresponding to the publisher-assigned user identifier obtained from the one or more database proprietors, the media access data, and the publisher-assigned user identifier;
generating, using the trained machine learning model and based on the input, probability values for corresponding audience demographics in a household composition corresponding to the publisher-assigned user identifier, the probability values indicative of likelihoods that corresponding ones of the audience demographics are accessing the connected television media;
determining a person-level characteristic based on the probability values of the audience demographics, the person-level characteristic corresponding to an audience member of the connected television media;
assigning media features extracted from the media access data to the person-level characteristic;
mapping the publisher-assigned user identifier to the person-level characteristic corresponding to the audience member of the connected television media; and
transmitting the publisher-assigned user identifier and the person-level characteristic to a publisher of the connected television media.
12. The non-transitory machine readable storage medium of claim 11 , wherein the operations further comprise:
identifying the household composition based on the user identifier, the household composition indicative of one or more demographic categories that use the media access device.
13. The non-transitory machine readable storage medium of claim 11 , wherein the person-level characteristic includes at least one of an age or a gender.
14. The non-transitory machine readable storage medium of claim 11 , wherein the operations further comprise:
analyzing historical media access data to determine the household composition associated with the media access device, the historical media access data generated during previous media access sessions presented by the media access device.
15. The non-transitory machine readable storage medium of claim 11 , wherein the operations further comprise:
generating a report including the audience member of the connected television media, the report to provide substantially real-time feedback of the audience member to an advertiser.
16. The non-transitory machine readable storage medium of claim 11 , wherein the operations further comprise:
obtaining an application programming interface call from the publisher of the connected television media, the application programming interface call indicative of second instructions to identify the audience member of the connected television media in substantially real-time.
17. The non-transitory machine readable storage medium of claim 11 , wherein the operations further comprise:
determining the person-level characteristic based on the person-level characteristic corresponding to a largest one of the probability values.
18. The non-transitory machine readable storage medium of claim 11 , wherein the operations further comprise:
generating, using a neural network, scores for demographic categories based on historical media access data; and
comparing the scores to a threshold to determine the household composition.
19. A method comprising:
obtaining media access data corresponding to connected television media and a user identifier corresponding to a media access device, wherein the user identifier comprises a publisher-assigned user identifier;
obtaining third-party data corresponding to the publisher-assigned user identifier from one or more database proprietors, the third-party data comprising information associated with a household composition corresponding to the publisher-assigned user identifier;
providing, as an input to a trained machine learning model, the third-party data corresponding to the publisher-assigned user identifier obtained from the one or more database proprietors, the media access data, and the publisher-assigned user identifier;
generating, using the trained machine learning model and based on the input, probability values for corresponding audience demographics in the household composition corresponding to the publisher-assigned user identifier, the probability values indicative of likelihoods that corresponding ones of the audience demographics are accessing the connected television media;
determining a person-level characteristic based on the probability values of the audience demographics, the person-level characteristic corresponding to an audience member of the connected television media;
assigning media features extracted from the media access data to the person-level characteristic;
mapping the publisher-assigned user identifier to the person-level characteristic corresponding to the audience member of the connected television media; and
transmitting the publisher-assigned user identifier and the person-level characteristic to a publisher of the connected television media.
20. The method of claim 19 , further comprising:
identifying the household composition based on the user identifier, the household composition indicative of one or more demographic categories that use the media access device.