Audience Metrics Matrix Imputation for Cookie-Limited Demographics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing audience measurement techniques face challenges in accurately determining audience demographics and sizes due to reliance on third-party cookies, which may be limited or unavailable, and panel data coverage is insufficient, leading to incomplete and biased audience metrics.
Innovation Solution
Implement a recommender system, such as a deep factorization machine, to impute missing demographic data using sparse matrix factorization, leveraging panel and census data to generate accurate audience metrics by normalizing and transforming data to maintain logical relationships and impute missing values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If panel data is used to measure audience demographics, then demographic information can be collected, but panel data coverage is insufficient leading to incomplete and biased audience metrics
Solution Approach 1:
The patent combines panel data with census data and recommender system predictions to create a hybrid measurement approach. This merging allows the system to leverage the demographic accuracy of panel data while compensating for its limited coverage using broader census data and predictive modeling, thereby resolving the contradiction between measurement precision and data quantity.
Solution Approach 2:
The recommender system acts as an intermediary between limited panel data and comprehensive audience measurement. It uses collaborative filtering algorithms to infer demographic characteristics for uncaptured audiences based on patterns from panel data, effectively bridging the coverage gap without sacrificing demographic accuracy.
2Loss of information
If third-party cookies are used for audience measurement, then audience data can be tracked, but cookies may be limited or unavailable leading to incomplete metrics
Solution Approach 1:
The patent extracts the core functionality of audience tracking from dependency on third-party cookies and implements it through alternative mechanisms including panel enrollment, device identification, and recommender system modeling. This extraction eliminates the reliability issue of cookie availability while maintaining audience tracking capabilities.
Solution Approach 2:
The system changes the fundamental parameters of audience identification from cookie-based tracking to multiple alternative identification methods including panelist enrollment status, device identifiers, and inferred demographic profiles. This parameter change makes the measurement system reliable regardless of cookie availability.
3Quantity of substance
If matrix factorization is used to impute missing data, then comprehensive audience metrics can be generated, but data transformation and normalization are required increasing processing complexity
Solution Approach 1:
The patent applies data normalization and transformation procedures before executing matrix factorization. By performing these preliminary actions, the system prepares the data in an optimal format that simplifies the subsequent imputation process and ensures consistent results, thereby managing processing complexity while achieving comprehensive metrics.
Data Source
AI summary
An example apparatus includes audience metrics collecting circuitry to access first audience metrics from a server, and access second audience metrics from the server, matrix building circuitry to build a matrix of the first audience metrics and the second audience metrics, missing values of the matrix corresponding to the second audience metrics, data transforming circuitry to transform the first audience metrics and the second audience metrics in the matrix, missing value calculating circuitry to determine imputed transformed values of the missing values using a recommender system, and the data transforming circuitry to recover imputed values of the missing values based on the imputed transformed values.