Person-Level Viewership Estimation via Household Log Segmentation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for estimating viewership among target audiences, such as active and household viewing logs, face limitations in providing accurate guarantees for person-level characteristics due to small sample sizes and lack of individual identification in household logs, making it difficult for broadcasters to offer reliable guarantees for specific demographics like males aged 18-49.
Innovation Solution
A computer-implemented method that assigns household viewing logs to buckets based on person-level data, generates a viewing behavior model using processors, and computes viewership among target audiences with specific characteristics, leveraging census data to ensure accuracy and reliability across a wide range of audiences.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If active viewing logs are used to estimate viewership, then person-level viewing identification is achieved, but sample size is too small to generate reliable estimates for large broadcast areas
Solution Approach 1:
The patent combines active viewing logs (which have person-level identification but small sample sizes) with household viewing logs (which have large sample sizes but lack person-level identification) by merging them into a unified data structure. This allows the system to leverage both the precision of active logs and the quantity of household logs to achieve reliable viewership estimates for large broadcast areas.
Solution Approach 2:
The patent introduces an intermediary data structure that bridges active viewing logs and household viewing logs. This intermediary structure enables the system to associate person-level characteristics with viewing behavior while maintaining the large sample size of household logs, thereby resolving the contradiction between measurement precision and quantity of substance.
2Quantity of substance
If household viewing logs are used to estimate viewership, then sample size is large enough for reliable estimates, but person-level identification is lost
Solution Approach 1:
The patent merges household viewing logs with active viewing logs to recover person-level identification while maintaining the large sample size of household logs. The unified data structure allows the system to attribute viewing behavior to specific persons within households, thereby achieving both large sample size and person-level precision.
Solution Approach 2:
The patent segments the viewing log data into distinct components: household-level viewing information and person-level characteristics. By segmenting the data this way, the system can analyze viewing behavior at both the household level (for large sample sizes) and the person level (for precise identification), resolving the contradiction between quantity and precision.
3Extent of automation
If household viewing logs are used, then automatic data collection is achieved, but inability to distinguish between persons in multi-person households prevents reliable target audience guarantees
Solution Approach 1:
The patent combines automatic household viewing log data with person-level characteristics from active viewing logs. This merging allows the system to maintain automatic data collection while adding the ability to distinguish between persons in households, thereby enabling reliable target audience guarantees for demographics like males aged 18-49.
Solution Approach 2:
The patent introduces an intermediary data structure that bridges automatic household viewing logs with person-level identification. This intermediary structure enables the system to automatically collect household viewing data while simultaneously identifying which persons in the household are watching, thereby resolving the contradiction between automation and reliability.
Data Source
AI summary
In one embodiment, a viewing behavior subsystem computes estimated viewership among target audiences at a person-level based on a household viewing logs and person-level data. A viewing behavior subsystem distributes the household viewing logs across buckets based on the sizes of the households and sets of person-level characteristics that are associated with the persons within the households. The viewing behavior subsystem generates a model for viewing behaviors associated with different sets of person-level characteristics based on the buckets. Subsequently, the viewing behavior subsystem estimates viewership among a target audience that is associated with one or more of the sets of person-level characteristics based on the model and census data. Advantageously, by combining household viewing logs for a relatively large number of households with person-level data, the viewing behavior subsystem enables accurate estimations of viewership among target audiences that are distinguished by person-level characteristics.