Probabilistic Media Viewing Metrics from Weighted Panelist Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for determining media viewing metrics, such as television ratings and shares, are inaccurate due to uncertainties in panelist viewing behavior data, which are not adequately addressed by current probabilistic simulations that often require extensive processing resources and limited scenario analysis.
Innovation Solution
A system and method that utilizes algorithms to account for all possible viewing scenarios by assigning sampling weights and probabilities to panelists, reducing the need for repetitive simulations, and calculating expected ratings and shares with improved accuracy and efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If probabilistic simulations are used to account for panelist viewing behavior uncertainties, then measurement precision of viewing metrics is improved, but device complexity and processing resources increase
Solution Approach 1:
The patent segments the probabilistic simulation process into distinct components: (1) receiving viewing data from multiple panelists, (2) determining probabilities for each panelist's viewing behavior, (3) generating multiple simulated viewing scenarios, (4) aggregating results across scenarios to produce expected viewing metrics. This segmentation allows each component to be processed independently and efficiently, reducing overall system complexity while maintaining measurement precision.
Solution Approach 2:
The patent performs preliminary actions by pre-determining probabilities for each panelist's viewing behavior before generating simulation scenarios. By calculating these probabilities in advance based on available viewing data, the system avoids complex real-time computations during scenario generation, thereby reducing processing complexity while preserving accuracy.
2Measurement precision
If extensive probabilistic simulations are performed to improve viewing metrics accuracy, then measurement precision is improved, but productivity and processing efficiency deteriorate
Solution Approach 1:
The patent applies partial action by generating aĉé number of simulation scenarios (e.g., 10-100 scenarios) rather than exhaustively simulating all possible viewing combinations. This partial simulation approach provides sufficiently accurate expected viewing metrics while maintaining high processing efficiency, avoiding the computational burden of complete enumeration.
Solution Approach 2:
The patent changes the parameter of simulation scenario quantity from potentially infinite to a manageable finite number. By adjusting this parameter and using statistical aggregation methods, the system achieves accurate viewing metrics with reduced computational requirements, thereby improving productivity without sacrificing measurement precision.
3Measurement precision
If repetitive simulations are conducted to account for all viewing scenarios, then measurement precision is improved, but loss of time increases
Solution Approach 1:
The patent implements continuous useful action by processing multiple simulation scenarios in parallel and continuously aggregating results to compute expected viewing metrics. This continuous processing approach eliminates idle time between simulations and efficiently utilizes computational resources, reducing total computation time while maintaining accuracy.
Solution Approach 2:
By pre-determining panelist viewing probabilities before simulation, the system eliminates the need for repeated probability calculations across scenarios. This preliminary action significantly reduces computation time while preserving the accuracy benefits of multiple scenario simulations.
Data Source
AI summary
Methods and apparatus to determine probabilistic media viewing metrics are disclosed herein. An example apparatus includes memory including machine reachable instructions; and processor circuitry to execute the instructions to calculate a first probability for respective ones of a plurality of panelists as having viewed media based on viewing data, the viewing data including incomplete viewing data for one or more of the panelists relative to the media; identify respective ones of a plurality of panelists as included in a demographic subgroup based on demographic data for the panelists; assign a sampling weight to the respective ones of the plurality of panelists based on the demographic data; and calculate a second probability of the demographic subgroup having viewed the media based on the first probabilities and the sampling weights for the respective ones of the plurality of panelists in the demographic subgroup.