Methods and apparatus to generate audience metrics using third-party privacy-protected cloud environments

A non-transitory computer readable medium comprising instructions that, when executed, cause a processor of a database proprietor computing system of a first cloud-based environment of a database proprietor to perform operations, wherein the database proprietor computing system is connected to an audience measurement computing system of a second cloud-based environment of an audience measurement entity, separate from the first cloud-based environment, via an application programming interface, which facilitates the database proprietor computing system using panel data from the audience measurement entity to perform demographic correction for impressions of media on websites of the database proprietor that are collected by the database proprietor computing system and provide the corrected impressions to the audience measurement computing system without the audience measurement computing system itself using tags of the audience measurement entity embedded in that media to collect the impressions, the database proprietor computing system further comprising:

• a network interface, and
• a privacy-protected datastore of the audience measurement entity, wherein the privacy-protected datastore is located within the first cloud-based environment and remote from the audience measurement computing system,
• the operations comprising:
• obtaining database proprietor impression data associated with multiple user accounts established by registered users of the database proprietor, the database proprietor impression data logged based on network communications from client devices of the registered users that accessed the media via the user accounts, the database proprietor impression data having a computer-generated error in which self-reported demographics of a user to which a set of impressions of the media are attributed do not correspond to actual demographics of the user;
• retrieving the panel data from the audience measurement computing system via the network interface and via the application programming interface, the panel data comprising media impression data and corresponding demographic data for panelists of the audience measurement entity that were exposed to the media;
• merging the panel data with the database proprietor impression data to generate merged data;
• based on the merged data, generating a demographic correction model, wherein the demographic correction model is configured to predict whether impressions of the database proprietor impression data correspond to a demographic;
• correcting the computer-generated error by (i) generating predicted demographics of the user by applying the demographic correction model to the set of impressions and (ii) changing the self-reported demographics attributed to the set of impressions to the predicted demographics, wherein the predicted demographics comprises the demographic;
• and
• storing the set of impressions having the attributed predicted demographics in the privacy-protected datastore,
• wherein the audience measurement computing system is configured to:
• collect the panel data by communicating with audience measurement meters of panelist households,
• provide the panel data to the database proprietor computing system via the application programming interface,
• retrieve the set of impressions and the attributed predicted demographics from the privacy-protected datastore, and
• generate a calibrated set of impression data based on the set of impressions and the attributed predicted demographics.

A method performed by database proprietor computing system of a first cloud-based environment of a database proprietor, wherein the database proprietor computing system is connected to an audience measurement computing system of a second cloud-based environment of an audience measurement entity, separate from the first cloud-based environment, via an application programming interface, which facilitates the database proprietor computing system using panel data from the audience measurement entity to perform demographic correction for impressions of media on websites of the database proprietor that are collected by the database proprietor computing system and provide the corrected impressions to the audience measurement computing system without the audience measurement computing system itself using tags of the audience measurement entity embedded in that media to collect the impressions, the database proprietor computing system comprising:

• a network interface,
• a privacy-protected datastore of the audience measurement entity, wherein the privacy-protected datastore is located within the first cloud-based environment and remote from the audience measurement computing system,
• a processor, and
• a memory,
• the method comprising:
• obtaining database proprietor impression data associated with multiple user accounts established by registered users of the database proprietor, the database proprietor impression data logged based on network communications from client devices of the registered users that accessed the media via the user accounts, the database proprietor impression data having a computer-generated error in which self-reported demographics of a user to which a set of impressions of the media are attributed do not correspond to actual demographics of the user;
• retrieving the panel data from the audience measurement computing system via the network interface and via the application programming interface, the panel data comprising media impression data and corresponding demographic data for panelists of the audience measurement entity that were exposed to the media;
• merging the panel data with the database proprietor impression data to generate merged data;
• based on the merged data, generating a demographic correction model, wherein the demographic correction model is configured to predict whether impressions of the database proprietor impression data correspond to a demographic;
• correcting the computer-generated error by (i) generating predicted demographics of the user by applying the demographic correction model to the set of impressions and (ii) changing the self-reported demographics attributed to the set of impressions to the predicted demographics, wherein the predicted demographics comprises the demographic;
• and
• storing the set of impressions having the attributed predicted demographics in the privacy-protected datastore,
• wherein the audience measurement computing system is configured to:
• collect the panel data by communicating with audience measurement meters of panelist households,
• provide the panel data to the database proprietor computing system via the application programming interface,
• retrieve the set of impressions and the attributed predicted demographics from the privacy-protected datastore, and
• generate a calibrated set of impression data based on the set of impressions and the attributed predicted demographics.