Methods and apparatus to determine unified entity weights for media measurement

A non-transitory machine readable storage medium comprising instructions that, when executed by an audience measurement computing system comprising:

• i) a semiconductor microprocessor device comprising a plurality of cores, wherein each of the cores is configured to operate independently, ii) a Field Programmable Gate Array (FPGA) comprising an array of logic gates, a plurality of configurable interconnections, and a storage circuitry, wherein the storage circuitry is distributed among the array of logic gates so as to increase execution speed, iii) a interface circuitry, and iv) a memory, cause the system to perform operations comprising:
• obtaining, via one or more network communications, entity data for a plurality of entities by:
• obtaining, by a first sub-array of logic gates of the FPGA and via the one or more network communications, audience measurement data from meter devices disposed at a plurality of media exposure measurement locations,
• storing, by the storage circuitry of the FPGA, a first result of the obtaining of the audience measurement data from meter devices,
• obtaining, by the second sub-array of logic gates the FPGA and via the one or more network communications, return path data from media devices disposed at the plurality of media exposure measurement locations, and
• storing, by the storage circuitry of the FPGA, a second result of the obtaining of return path data from the media devices;
• wherein each entity of the plurality of entities comprises:
• a person, or a household representing a media exposure measurement location, and wherein, for each entity of the plurality of entities:
• i) the entity data represents a period of time comprising a plurality of time segments, and ii) the entity data comprises:
• an entity identifier for the entity and, for each time segment in the period of time:
• an adaptable weight of the entity and a panel status of the entity;
• determining, at each iteration of a plurality of iterations executed in parallel by the plurality of cores of the semiconductor microprocessor device, a unified weight for each respective entity of the plurality of entities, comprising, at each iteration:
• averaging, by the respective core, the adaptable weights based in the panel statuses to determine the unified weight of the respective entity for the period of time;
• and
• outputting, by the plurality of cores of the semiconductor microprocessor device and the interface circuitry, the unified weights of the plurality of entities to an external device.

A method performed by an audience measurement computing system comprising:

• i) a semiconductor microprocessor device comprising a plurality of cores, wherein each of the cores is configured to operate independently, ii) a Field Programmable Gate Array (FPGA) comprising an array of logic gates, a plurality of configurable interconnections, and a storage circuitry, wherein the storage circuitry is distributed among the array of logic gates so as to increase execution speed, iii) a interface circuitry, and iv) a memory, the method comprising:
• obtaining, via one or more network communications, entity data for a plurality of entities by:
• obtaining, by a first sub-array of logic gates of the FPGA and via the one or more network communications, audience measurement data from meter devices disposed at a plurality of media exposure measurement locations,
• storing, by the storage circuitry of the FPGA, a first result of the obtaining of the audience measurement data from meter devices,
• obtaining, a second sub-array of logic gates of the FPGA and via the one or more network communications, return path data from media devices disposed at the plurality of media exposure measurement locations, and
• storing, by the storage circuitry of the FPGA, a second result of the obtaining of return path data from the media devices;
• wherein each entity of the plurality of entities comprises:
• a person, or a household representing a media exposure measurement location, and wherein, for each entity of the plurality of entities:
• i) the entity data represents a period of time comprising a plurality of time segments, and ii) the entity data comprises:
• an entity identifier for the entity and, for each time segment in the period of time:
• an adaptable weight of the entity and a panel status of the entity;
• and
• determining, at each iteration of a plurality of iterations executed in parallel by the plurality of cores of the semiconductor microprocessor device, a unified weight for each respective entity of the plurality of entities, comprising, at each iteration:
• averaging, by the respective core, the adaptable weights based on the panel statuses to determine the unified weight of the respective entity for the period of time, and
• outputting, by the plurality of cores of the semiconductor or microprocessor device and the interface circuitry, the unified weights of the plurality of entities to an external device.