Audience Deduplication via Bayesian Inference and Intermediary Matching
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for monitoring user access to internet-accessible media, such as digital television and digital advertisements, face challenges like over-counting and under-counting due to server log tampering and caching issues, and the limitations of third-party cookie tracking, which hinder accurate audience measurement and deduplication across multiple platforms.
Innovation Solution
The implementation of a system that uses media deduplication circuitry to match panelists with database proprietor accounts using PII variables, employing Bayesian inference and iterative proportional fitting to determine deduplicated audience sizes across platforms, and leveraging database proprietors to collect impression data without revealing personally identifiable information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If third-party cookie tracking is used to collect impression data across platforms, then audience measurement coverage is improved, but measurement precision deteriorates due to over-counting and under-counting issues
Solution Approach 1:
The patent introduces database proprietors as intermediary entities that collect and manage impression data without revealing personally identifiable information. These intermediaries aggregate data from multiple platforms and provide it to the audience measurement entity, eliminating the need for direct third-party cookie tracking while maintaining measurement coverage and improving accuracy through centralized deduplication processes
Solution Approach 2:
The patent extracts and removes duplicate impressions from the collected data by identifying and eliminating redundant records across platforms. This extraction process separates unique audience members from duplicate counts, directly improving measurement precision while preserving the comprehensive coverage achieved through multi-platform data collection
2Measurement precision
If PII data is collected for matching panelists with accounts, then measurement precision is improved, but loss of information deteriorates due to privacy concerns
Solution Approach 1:
The patent extracts only the necessary personally identifiable information variables required for matching panelists with database proprietor accounts, while excluding unnecessary PII data. This selective extraction maintains measurement precision by preserving essential matching criteria while minimizing privacy information loss through targeted data collection
Solution Approach 2:
The patent uses database proprietors as intermediaries that handle PII data collection and matching processes. These intermediaries manage the sensitive information securely, performing the matching function while protecting privacy by not requiring the audience measurement entity to directly access or store comprehensive PII data sets
Data Source
AI summary
Methods, apparatus, and articles of manufacture to deduplicate audiences across media platforms are disclosed. An example apparatus includes memory; and processor circuitry to execute the instructions to: generate a match panel by matching panelists with database proprietor accounts based on matching information; generate respondent-level data from the match panel by combining first media exposure data corresponding to panelists associated with the match panel and second media exposure data corresponding to the database proprietor accounts associated with the match panel, the first and second media exposure data corresponding to a media item; determine a probability distribution corresponding to observed deduplication audience size data, the observed deduplication audience size data based on the respondent-level data of the match panel; perform iterative proportional fitting on an output probability corresponding to the probability distribution; and determine a deduplicated total audience size for the media item based on a result of the iterative proportional fitting.