Media Consumption Imputation for Localized Event Bias
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audience measurement systems face imputation errors due to localized events, which skew audience configurations and lead to inaccurate advertising pricing by mismatching tuning and viewing panelists across different designated market areas (DMAs).
Innovation Solution
Implement a localized event adjustment process that identifies heavily exposed data sets, performs probability adjustments, and expands the learning household pool to match tuning panelists with viewing panelists, reducing imputation errors through techniques like Localized Event Recipient Cutback (LERC) and Localized Event Donor Cutback (LEDC) data matching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If panelist data is collected and matched across different DMAs using traditional imputation methods, then audience measurement coverage is improved, but imputation errors increase due to localized events skewing audience configurations
Solution Approach 1:
The system performs preliminary identification of localized events (such as sports games, news events, or other media content that generates concentrated viewing) before conducting the imputation process. By detecting these events in advance using media content analysis and viewing pattern recognition, the system can apply specific adjustments to panelist data matching, preventing imputation errors from occurring in the first place rather than correcting them afterward.
Solution Approach 2:
The system applies different imputation strategies based on local conditions in different DMAs. When localized events are detected in specific DMAs, the system adjusts the imputation process for those particular regions while maintaining standard procedures in other areas. This allows the system to maintain high measurement precision across all DMAs while applying targeted corrections only where needed, thereby improving overall reliability without sacrificing coverage.
2Productivity
If tuning panelists are matched with viewing panelists using standard imputation techniques, then audience configuration estimates are generated, but advertising pricing accuracy deteriorates due to mismatches caused by localized events
Solution Approach 1:
The system incorporates feedback loops that continuously monitor the matching between tuning panelists and viewing panelists. When localized events are detected, the system receives feedback about discrepancies in audience configurations and automatically adjusts the imputation process. This feedback mechanism ensures that advertising pricing accuracy is maintained by dynamically correcting mismatches that arise from concentrated viewing patterns during localized events.
Solution Approach 2:
The system dynamically changes imputation parameters based on the detection of localized events. When such events are identified, the system adjusts weighting factors, matching thresholds, and statistical models used in the imputation process. These parameter changes allow the system to maintain productivity by continuing to generate audience configuration estimates efficiently, while simultaneously improving advertising pricing accuracy by adapting to the altered viewing patterns caused by localized events.
Data Source
AI summary
Methods, apparatus, systems and articles of manufacture to impute media consumption behavior are disclosed. An example system includes one or more media meters to obtain tuning data, one or more people meters to obtain viewing data, and one or more servers to, in response to a determination that a difference satisfies a first threshold, determine that a first subset of the tuning data associated with first panelist households having tuned to first media in a first area exhibits local bias, determine that a second subset of the viewing data associated with second panelist households having viewed the first media in the second area represents heavy viewing, and impute the second subset of the viewing data for the first subset of the tuning data in response to the second subset of the viewing data representing heavy viewing.