Synthetic Behavioral Profiles for Privacy-Preserving Audience Measurement
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The increasing concern for digital privacy has led to limitations on data granularity and access to individual profiles due to regulations and breaches, making it challenging to provide detailed behavioral profiles without compromising user privacy.
Innovation Solution
Utilizing neural networks to generate synthetic behavioral profiles through census network data and tag-less partner integration, breaking the linkage to personal identifiers by creating atomic IDs that represent expected web traffic.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If individual profile data is provided to satisfy data granularity requirements, then measurement precision is improved, but privacy protection deteriorates
Solution Approach 1:
The patent creates synthetic personas that copy and replicate the behavioral patterns of real users without using actual user data. The neural network generates fictional user profiles that mirror real user behaviors, allowing measurement precision to be maintained while privacy is protected since the personas are synthetic copies rather than real individual data
Solution Approach 2:
The patent introduces synthetic personas as an intermediary between real users and the measurement system. These personas act as mediators that represent user behavior patterns without directly exposing individual user identities, allowing the system to access detailed behavioral data while maintaining user privacy through this intermediate layer
2Object-affected harmful factors
If identity obfuscation is applied to protect user privacy, then privacy protection is improved, but data reliability deteriorates
Solution Approach 1:
Instead of obfuscating real user data, the system creates synthetic copies (personas) that replicate user behavior patterns. This approach maintains data reliability because the personas are generated to accurately reflect real user behaviors through neural network training, while privacy is protected since the underlying real user identities are not exposed or linked to the data
3Measurement precision
If real user data is used to generate profiles, then measurement precision is improved, but loss of information increases due to privacy regulations and data breaches
Solution Approach 1:
The system uses neural networks to learn from real user data and generate synthetic personas that copy user behavior patterns. Once trained, the model can generate new personas without accessing or storing the original real user data, maintaining measurement precision through accurate behavioral replication while eliminating the risk of data breaches and privacy violations associated with storing and accessing real user information
Data Source
AI summary
Online consumption data may be secured by receiving data, clustering elements of the received data into clusters, measuring an anonymity of each cluster based on entropy, determining that the anonymity of a first cluster does not satisfy a predetermined threshold, modifying the first cluster, measuring an anonymity of the modified first cluster based on entropy, determining that the anonymity of the modified first cluster does satisfy the predetermined threshold, and not further modifying the modified first cluster.