Audience Demographic Estimation via Cosine Similarity
Solution Overview
Problem
Current methods for determining demographic compositions of television audiences are impractical and costly, especially for unrated programs, as they often require extensive sampling and may not provide reliable demographic data due to insufficient viewer numbers.
Innovation Solution
The method involves accessing ratings data for rated channel time blocks, determining similarity measures with unrated blocks using segment data, and estimating demographic characteristics through a process that leverages the demographic characteristics of similar rated blocks, thereby reducing costs and improving data availability for unrated programs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If panels or surveys are used to estimate viewer demographics, then demographic data can be obtained, but it is impractical to sample enough households to determine audience composition for every television program
Solution Approach 1:
The patent uses segment characteristics as an intermediary to bridge unrated programs and rated programs. By comparing segment data (which is available for all programs) with demographic data (which is only available for rated programs), the system can estimate demographics for unrated programs without direct sampling, thus resolving the contradiction between data accuracy and collection efficiency
Solution Approach 2:
The patent creates a copy of demographic information from rated programs and applies it to unrated programs through similarity matching. Instead of collecting original demographic data through expensive panels for every program, the system copies and adapts demographic profiles from similar rated programs, achieving cost-effective demographic estimation
2Measurement precision
If extensive sampling is conducted to determine demographic compositions, then more accurate demographic data can be obtained, but the cost increases significantly
Solution Approach 1:
The system copies demographic profiles from rated programs to unrated programs by matching segment characteristics. This avoids the need for expensive extensive sampling for every program while maintaining reasonable demographic accuracy through the use of similar program profiles
Solution Approach 2:
The patent makes segment characteristics serve multiple functions: they are used both for program classification and as a basis for demographic estimation. This multi-functionality allows the same data to drive both program organization and demographic inference, reducing the need for separate expensive sampling operations
3Loss of information
If demographic data is collected for every program through traditional methods, then complete demographic coverage is achieved, but the complexity and time required increases
Solution Approach 1:
The patent performs preliminary matching of segment characteristics between unrated and rated programs before demographic estimation. By pre-organizing and comparing segment data, the system quickly identifies suitable rated programs to serve as demographic sources, avoiding time-consuming direct sampling for each unrated program
Solution Approach 2:
Segment characteristics serve as an intermediary layer that enables rapid matching between unrated and rated programs. This intermediary data structure allows the system to efficiently bridge the gap without requiring time-intensive direct demographic measurement for every program
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for estimating demographic compositions of television audiences from audience similarities. In an aspect, pairwise cosine similarities of segment of viewers of rated and unrated television programs are determined, demographic information of rated programs is attributed to the unrated programs from the cosine similarity measures and demographic information for the rated programs.