Cooperative Synthetic Identity Manager for Privacy-Preserving Analytics
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for analyzing patterns of behavior and social relationships across organizations often expose personally identifiable information, posing a threat to privacy and requiring consistent data masking across datasets to preserve identity relationships.
Innovation Solution
The creation and management of cooperative synthetic identities, which replace personally identifiable information with synthetic data, allowing for consistent data masking and cross-organizational analytics while protecting privacy, and can be associated with specific cases and expired on a certain period.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of information
If personally identifiable information is used for cross-organizational analytics, then behavioral and social pattern analysis can be performed, but privacy is exposed and liability increases
Solution Approach 1:
The patent creates synthetic identity copies that replicate the behavioral and social relationship patterns of original identities without exposing real personally identifiable information. These synthetic identities serve as safe proxies for analytics, allowing organizations to analyze patterns while protecting privacy and reducing liability associated with handling real PII.
Solution Approach 2:
The synthetic identity acts as an intermediary between the original identity and the analytics process. Instead of directly analyzing real PII, the system uses synthetic identities as a mediating layer that preserves pattern information while eliminating privacy risks and liability concerns associated with direct PII analysis.
2Object-affected harmful factors
If data masking is applied to protect privacy, then personally identifiable information is protected, but consistent data masking across datasets becomes difficult to maintain
Solution Approach 1:
Instead of applying complex masking rules to protect PII, the system creates synthetic identity copies that inherently protect privacy while maintaining consistent relationships across datasets. The synthetic identities are generated with consistent masking applied uniformly, eliminating the need for complex ongoing management of masking consistency.
Solution Approach 2:
The system transforms the data representation by changing from real PII parameters to synthetic identity parameters. This parameter transformation maintains the structural relationships and patterns needed for analytics while automatically ensuring consistent protection across all datasets, as the synthetic identities are generated with uniform privacy protections built-in.
3Object-affected harmful factors
If synthetic identities are created for each request, then privacy is protected, but system complexity and processing time increase
Solution Approach 1:
The system performs preliminary actions by pre-generating and storing synthetic identities in a cache before they are needed for analytics requests. When a request comes in, the system checks the cache for existing synthetic identities and reuses them, avoiding the need to create new synthetic identities for each request. This preliminary preparation reduces both system complexity and processing time while maintaining privacy protection.
Solution Approach 2:
The synthetic identity system is designed to be universal and reusable across multiple requests and organizations. Once a synthetic identity is created, it can be used repeatedly for different analytics queries and shared across organizational boundaries, eliminating the need to create unique synthetic identities for each request and reducing overall system complexity.
4Measurement precision
If real identities are used for analytics, then accurate behavioral patterns can be analyzed, but liability and privacy risks increase
Solution Approach 1:
The system creates synthetic identity copies that preserve the behavioral patterns and social relationships of original identities with sufficient accuracy for analytics purposes. These copies maintain the essential characteristics needed for pattern recognition while completely eliminating the privacy and liability risks associated with using real identities, thus achieving both measurement precision and reliability.
Data Source
AI summary
According to another embodiment of the disclosure, a method includes determining a first original identifier. The original identifier uniquely identifies a first original identity. The method also includes sending a request to a cooperative synthetic identity manager to create a first cooperative synthetic identity case for the first original identity. The method further includes receiving, from the cooperative synthetic identity manager, a first case identifier that uniquely identifies the first cooperative synthetic identity case. The method also includes requesting, from a first data entity, first cooperative synthetic identity information for the first original identity. The first cooperative synthetic identity information request comprises the first case identifier and the first original identifier. The method also includes receiving, from the first data entity, the first cooperative synthetic identity information.