Cohort Feature Data for Privacy-Preserving Ad Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional approaches fail to effectively measure the impact of advertisements on user behavior while protecting individual user privacy, as they require sensitive data controllers to share specific user information with modelers, compromising privacy.
Innovation Solution
The system generates and transmits aggregated cohort feature data instead of individual user data, allowing modelers to make probabilistic determinations about user behavior without accessing sensitive information, ensuring user privacy by grouping users into diverse cohorts.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sensitive user data is shared with modelers to enable accurate measurement of advertisement effectiveness, then measurement precision is improved, but user privacy is compromised
Solution Approach 1:
The patent introduces cohort feature data as an intermediary representation that enables modelers to measure advertisement effectiveness without direct access to sensitive user data. The cohort feature data aggregates user information into group-level statistics (e.g., conversion rates, engagement metrics) that preserve privacy while maintaining analytical utility. This mediator layer allows both parties to benefit: advertisers get measurement insights while users retain privacy control.
Solution Approach 2:
The patent transforms sensitive individual user data into aggregated cohort-level parameters and statistics. By changing the granularity and form of the data (from individual records to cohort summaries), the system maintains the ability to measure advertisement effectiveness while eliminating direct identification of individual users. The transformation preserves statistical properties needed for analysis while removing privacy risks.
2Object-affected harmful factors
If individual user data is protected to maintain privacy, then user privacy is improved, but the ability to measure advertisement impact deteriorates
Solution Approach 1:
The cohort feature data serves as a mediator that bridges the gap between privacy protection and measurement capability. It provides modelers with sufficient statistical information to assess advertisement impact while maintaining user privacy through aggregation. The intermediary structure enables both privacy preservation and effective measurement simultaneously.
Solution Approach 2:
The system changes the parameters from individual user attributes to cohort-level aggregated statistics. This parameter transformation maintains the essential information needed for measurement (such as conversion rates, engagement metrics) while changing the form of data to protect individual privacy. The aggregated parameters retain analytical value for advertising measurement.
3Reliability
If detailed user information is collected to improve modeling accuracy, then model reliability is improved, but data security requirements increase
Solution Approach 1:
The patent extracts only the necessary aggregated information from detailed user data to create cohort feature data. By taking out only the essential statistical properties needed for modeling (such as group-level behavior patterns) and leaving behind the sensitive individual details, the system maintains model reliability while reducing data security complexity. The extraction process retains analytical value while minimizing security risks.
Data Source
AI summary
Systems, methods, and non-transitory computer-readable media can generate individual feature data for each user of a plurality of users. A first cohort comprising a first plurality of users is generated, wherein the first plurality of users are selected from the plurality of users based on the individual feature data. A first set of cohort feature data associated with the first cohort is generated based on individual feature data for the first plurality of users. The first set of cohort feature data and a first set of cohort membership information are transmitted to a modeler. The first set of cohort membership information identifies each user of the plurality of users.