Aggregation-based methods for detection and correction of television viewership aberrations
View Patent ↗Various of the disclosed embodiments contemplate television viewing behavior data collected from a plurality of television set top boxes by using aggregation to detect an excess or a deficit in viewership for a group of television set top boxes. In some embodiments, a group of set top boxes can be associated with a particular television service provider, cable television head-end, or data warehouse. Additionally, some embodiments can clean television viewing behavior data by detecting and correcting aberrant viewership in a time series, e.g. based on a weekly or an approximately monthly frequency. In some embodiments, the aberrant viewership can be detected by calculating a minimum expected number of viewers for a day and comparing it to the actual number of households that reported viewers for that day.
1. A computer-implemented method for detecting and correcting a viewership aberration, the method comprising:
selecting a first dataset corresponding to a first plurality of set top boxes associated with a first service provider;
selecting a second dataset corresponding to a second plurality of set top boxes associated with at least a second service provider, the second service provider different from the first service provider, the second plurality of set top boxes different from the first plurality of set top boxes;
aggregating the first dataset across the first plurality of set top boxes to create a first viewing data pattern for a selected period of time;
aggregating the second dataset across the second plurality of set top boxes to create a second viewing data pattern for the selected period of time;
identifying a deviation in the first viewing data pattern relative to the second viewing data pattern using a criterion, wherein the criterion comprises a minimum viewership threshold; and
calculating a corrected viewership estimate based on the deviation.
2. The computer-implemented method of claim 1 , wherein identifying a deviation further comprises:
identifying a portion of the first viewing data pattern that is less than the threshold relative to the second viewing data pattern.
3. The computer-implemented method of claim 1 , wherein the criterion comprises a relationship between the first viewing data pattern and a regression model, wherein the regression model is calculated from historic viewing data weighted according to a periodically recurring time period.
4. The computer-implemented method of claim 1 , further comprising:
normalizing at least one of the first viewing data pattern and the second viewing data pattern so that the first viewing data pattern and the second viewing data pattern are comparable.
5. The computer-implemented method of claim 1 , further comprising:
calculating a normalization factor according to at least a relationship of a size associated with the first service provider compared to a size associated with the second service provider;
applying the normalization factor to the first viewing data pattern and the second viewing data pattern to make them comparable when the service providers have different customer footprints; and
if a deviation was determined, flagging the deviation for factoring into the calculation of the corrected viewership estimate.
6. A non-transitory computer-readable medium comprising instructions executable by at least one processor to perform a method comprising:
selecting a first dataset corresponding to a first plurality of set top boxes associated with a first service provider;
selecting a second dataset corresponding to a second plurality of set top boxes associated with at least a second service provider, the second service provider different from the first service provider, the second plurality of set top boxes different from the first plurality of set top boxes;
aggregating the first dataset across the first plurality of set top boxes to create a first viewing data pattern for a selected period of time;
aggregating the second dataset across the second plurality of set top boxes to create a second viewing data pattern for the selected period of time;
identifying, a deviation in the first viewing data pattern relative to the second viewing data pattern using a criterion, wherein the criterion comprises a minimum viewership threshold; and
calculating a corrected viewership estimate based on the deviation.
7. The non-transitory computer-readable medium of claim 6 , wherein identifying a deviation comprises:
identifying a portion of the first viewing data pattern that falls beneath the threshold relative to the second viewing data pattern.
8. The non-transitory computer-readable medium of claim 6 , wherein the criterion comprises a relationship between the first viewing data pattern and a regression model, wherein the regression model is calculated from historic viewing data weighted according to a periodically recurring time period.
9. The non-transitory computer-readable medium of claim 6 , the method further comprising:
normalizing at least one of the first viewing data pattern and the second viewing data pattern so that the first viewing data pattern and the second viewing data pattern are comparable.
10. The non-transitory computer-readable medium of claim 6 , the method further comprising:
calculating a normalization factor according to at least a relationship of a size associated with the first service provider compared to a size associated with the second service provider;
applying the normalization factor to the first viewing data pattern and the second viewing data pattern to make them comparable when the service providers have different customer footprints; and
if a deviation was determined, flagging the deviation for factoring into the calculation of the corrected viewership estimate.
11. A system for detecting and correcting a viewership aberration, the system comprising:
a storage comprising set top box data;
at least one processor;
a memory comprising instructions executable by the at least one processor to:
select a first dataset corresponding to a first plurality of set top boxes associated with a first service provider from the storage;
select a second dataset corresponding to a second plurality of set top boxes associated with at least a second service provider from the storage, the second service provider different from the first service provider, the second plurality of set top boxes different from the first plurality of set top boxes;
aggregate the first dataset across the first plurality of set top boxes to create a first viewing data pattern for a selected period of time;
aggregate the second dataset across the second plurality of set top boxes to create a second viewing data pattern for the selected period of time;
identify a deviation in the first viewing data pattern relative to the second viewing data pattern using a criterion, wherein the criterion comprises a minimum viewership threshold; and
calculate a corrected viewership estimate based on the deviation.
12. The system of claim 11 , wherein identifying a deviation comprises:
identifying a portion of the first viewing data pattern that falls beneath the threshold relative to the second viewing data pattern.
13. The system of claim 11 , wherein the criterion comprises a relationship between the first viewing data pattern and a regression model, wherein the regression model is calculated from historic viewing data weighted according to a periodically recurring time period.
14. The system of claim 11 , the instructions further configured, in conjunction with the at least one processor, to:
normalize at least one of the first viewing data pattern and the second viewing data pattern so that the first viewing data pattern and the second viewing data pattern are comparable.
15. The system of claim 11 , the instructions further executable by the at least one processor to:
calculate a normalization factor according to at least a relationship of a size associated with the first service provider compared to a size associated with the second service provider;
apply the normalization factor to the first viewing data pattern and the second viewing data pattern to make them comparable when the service providers have different customer footprints; and
if a deviation was determined, flag the deviation for factoring into the calculation of the corrected viewership estimate.