Granted Patent B2

US 12,407,898 · App. 17/666,312 · Granted Sep 2, 2025

Inventors: Michael Evan Anderson (Chicago, IL); Samantha M. Mowrer (San Francisco, CA); Robert Gottesman (Chicago, IL); Astha Jain (El Cerrito, CA)

Assignee: The Nielsen Company (US), LLC

H04N21/44222H04H60/31H04H60/37H04H60/45H04N21/6582

Abstract

Claims (65)

1. A computing system comprising:

a processor; and

a non-transitory computer-readable storage medium, having stored thereon program instructions that, upon execution by the processor, cause performance of a set of operations comprising:

training a tab status model using a training data set;

testing the tab status model with a subset of the training data set that was not used to train the tab status model to obtain a plurality of test output tab statuses;

comparing the plurality of test output tab statuses from the tab status model with corresponding tab statutes of the subset of the training data that was not used to train the tab status model;

based on the comparison, determining that an accuracy of the tab status model exceeds an accuracy threshold;

providing smart television data from a smart television as an input to the tab status model to generate an output,

wherein the smart television data is included in population data from a population of smart televisions;

the training data set, comprising, for each of multiple smart televisions, respective training data that is labeled as in-tab or out-of-tab based on whether the smart television lacked network connectivity for a threshold amount of time preventing transmission of additional smart television data to a server;

determining a tab status of the smart television based on the output of the tab status model;

removing the smart television data from the population data based on the tab status of the smart television being out-of-tab; and

crediting media in a report based on the population data after removing the smart television data from the population data.

2. The computing system of claim 1 , wherein the tab status is in-tab when the smart television transmits data to the server within the threshold amount of time; and wherein the tab status is out-of-tab when the smart television transmits data outside the threshold amount of time from an anticipated time of transmission.

3. The computing system of claim 1 , wherein determining the tab status comprises:

determining that the smart television is out-of-tab, if the output satisfies a threshold; and

determining that the smart television is in-tab, if the output does not satisfy the threshold.

4. The computing system of claim 1 , wherein the smart television is a second smart television; and wherein the set of operations further comprises:

identifying a first smart television that corresponds to a first panelist by comparing first data from the first smart television to second data from the first panelist; and

generating training data by labelling the first smart television as in-tab or out-of-tab based on the second data from the first panelist.

5. The computing system of claim 4 , wherein the first data includes at least one of a media identifier, a smart television identifier, a timestamp, tuning data, or disconnect data.

6. The computing system of claim 4 , wherein the second data is meter data indicative of media exposure of the first panelist.

7. The computing system of claim 4 , wherein identifying the first smart television corresponds to the first panelist comprises determining when a threshold amount of tuning data of the first data is consistent with the second data.

8. The computing system of claim 4 , wherein training the tab status model based partly on the training data set further comprises:

labeling the first smart television as out-of-tab when tuning data of the first data is inconsistent with meter data of the second data.

9. A non-transitory computer-readable storage medium, having stored thereon program instructions that, upon execution by a processor, cause performance of a set of operations comprising:

training a tab status model using a training data set;

testing the tab status model with a subset of the training data set that was not used to train the tab status model to obtain a plurality of test output tab statuses;

comparing the plurality of test output tab statuses from the tab status model with corresponding tab statutes of the subset of the training data that was not used to train the tab status model;

based on the comparison, determining that an accuracy of the tab status model exceeds an accuracy threshold;

providing smart television data from a smart television as an input to the tab status model to generate an output,

wherein the smart television data is included in population data from a population of smart televisions;

the training data set, comprising, for each of multiple smart televisions, respective training data that is labeled as in-tab or out-of-tab based on whether the smart television lacked network connectivity for a threshold amount of time preventing transmission of additional smart television data to a server;

determining a tab status of the smart television based on the output of the tab status model;

removing the smart television data from the population data based on the tab status of the smart television being out-of-tab; and

crediting media in a report based on the population data after removing the smart television data from the population data.

10. The non-transitory computer-readable storage medium of claim 9 , wherein the tab status is in-tab when the smart television is capable of transmitting data to the server.

11. The non-transitory computer-readable storage medium of claim 10 , wherein determining the tab status comprises:

determining that the smart television is out-of-tab, if the output satisfies a threshold; and

determining that the smart television is in-tab, if the output does not satisfy the threshold.

12. The non-transitory computer-readable storage medium of claim 9 , wherein the smart television is a second smart television, wherein the set of operations further comprises:

identifying a first smart television that corresponds to a first panelist by comparing first data from the first smart television to second data from the first panelist; and

generating training data by labelling the first smart television as in-tab or out-of-tab based on the second data from the first panelist.

13. The non-transitory computer-readable storage medium of claim 12 , wherein the first data includes at least one of a media identifier, a smart television identifier, a timestamp, tuning data, or disconnect data.

14. The non-transitory computer-readable storage medium of claim 12 , wherein the second data is meter data indicative of media exposure of the first panelist.

15. The non-transitory computer-readable storage medium of claim 12 , wherein identifying the first smart television corresponds to the first panelist comprises determining when a threshold amount of tuning data of the first data is consistent with the second data.

16. The non-transitory computer-readable storage medium of claim 12 , wherein training the tab status model based partly on the training data set further comprises: labeling the first smart television as out-of-tab when tuning data of the first data is inconsistent with meter data of the second data.

17. A method comprising:

training a tab status model using a training data set;

testing the tab status model with a subset of the training data set that was not used to train the tab status model to obtain a plurality of test output tab statuses;

comparing the plurality of test output tab statuses from the tab status model with corresponding tab statutes of the subset of the training data that was not used to train the tab status model;

based on the comparison, determining that an accuracy of the tab status model exceeds an accuracy threshold;

providing smart television data from a smart television as an input to the tab status model to generate an output,

wherein the smart television data is included in population data from a population of smart televisions;

the training data set, comprising, for each of multiple smart televisions, respective training data that is labeled as in-tab or out-of-tab based on whether the smart television lacked network connectivity for a threshold amount of time preventing transmission of additional smart television data to a server;

determining a tab status of the smart television based on the output of the tab status model;

discarding the smart television data from the population data based on the smart television being out-of-tab; and

crediting media in a report based on the population data after removing the smart television data from the population data.

18. The method of claim 17 , wherein the tab status is in-tab when the smart television is capable of transmitting data to the server or out-of-tab.

19. The method of claim 18 , wherein determining the tab status comprises:

determining that the smart television is out-of-tab, if the output satisfies a threshold; and

determining that the smart television is in-tab, if the output does not satisfy the threshold.

20. The method of claim 17 , wherein the smart television is a second smart television, further comprising:

identifying a first smart television that corresponds to a first panelist by comparing first data from the first smart television to second data from the first panelist; and

generating training data by labelling the first smart television as in-tab or out-of-tab based on the second data from the first panelist.

Assignments (4)

Continuity (2)

References Cited (24)