Contextual Session Stitching With Tunable Privacy Buckets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing user session continuity technologies, such as universal device identifiers and user fingerprints, fail to meet user privacy and security expectations due to exposure of sensitive information and potential misuse.
Innovation Solution
A tunable bucket identifier system is used to stitch user sessions across disjoint entities, balancing accuracy and privacy by ensuring the identifier is common enough to be shared but unique enough to maintain contextually relevant session continuity.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If universal device identifiers are used to facilitate session continuity, then session continuity is improved, but user privacy and security are worsened due to exposure of sensitive information
Solution Approach 1:
The patent extracts the identifying information from universal device identifiers and user fingerprints, replacing them with synthetic bucket identifiers that do not expose sensitive user data. The bucket identifiers are derived from non-sensitive attributes (device type, OS version, app version) to maintain identification capability while removing harmful privacy exposures.
Solution Approach 2:
The patent introduces bucket identifiers as an intermediary between the need for user identification and the requirement for privacy protection. These synthetic identifiers act as a mediator that enables session continuity across devices without directly exposing sensitive user information, thereby resolving the contradiction between reliability and privacy.
2Reliability
If user fingerprints are used to facilitate session continuity, then session continuity is improved, but user privacy and security are worsened due to potential misuse of user information
Solution Approach 1:
The patent removes sensitive user information from the identification process by extracting only the necessary non-sensitive attributes (device type, operating system version, application version) to generate bucket identifiers. This extraction eliminates the risk of misuse while preserving the ability to continue user sessions.
Solution Approach 2:
The patent changes the parameters used for identification from sensitive user fingerprints to non-sensitive version parameters. By transforming the identification basis from personal characteristics to aggregate version information, the system maintains session continuity functionality while eliminating the harmful potential for information misuse.
3Measurement precision
If a unique identifier is used to ensure accurate user identification, then identification accuracy is improved, but user privacy is worsened due to exposure of sensitive user data
Solution Approach 1:
The patent creates synthetic copies of user identification through bucket identifiers that replicate the functional capability of unique identifiers without containing actual sensitive user data. These copies are generated from aggregated version information and provide sufficient accuracy for session continuity while maintaining privacy by never storing or exposing real user identifiers.
Solution Approach 2:
The patent fundamentally changes the parameters used for identification from sensitive user-specific data to non-sensitive version parameters. This parameter transformation enables the system to achieve identification accuracy through aggregate statistics rather than individual user data, thereby resolving the contradiction between measurement precision and privacy protection.
Data Source
AI summary
Aspects of the disclosure are directed to stitching together user sessions across disjoint entities to facilitate session continuity while meeting accuracy and privacy thresholds. Stitching together the user sessions involves a tunable bucket identifier for one or more users. The bucket identifier can be tuned to be common enough to be shared by multiple users but unique enough to allow for stitching together a user session when contextually relevant.