Distributed Share Anonymization for Cross-Organization Data Analysis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In secure computing, cross-sectional analysis between organizations requires sharing personal information, necessitating consent from individuals, which is a challenge due to data confidentiality constraints.
Innovation Solution
A processing system that segments and distributes registration data to multiple servers, performing anonymization on shares to provide anonymization-processed information without revealing individual identities, enabling secure computation without personal information disclosure.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If personal information is shared for cross-sectional analysis between organizations, then data utility and analytical value are improved, but data confidentiality and privacy protection deteriorate
Solution Approach 1:
The patent segments personal information into multiple shares using secret sharing technology, distributing them across different servers. No single server holds the complete information, making it impossible to reconstruct original data without all shares. This enables cross-organizational analysis while maintaining confidentiality, as data is utility-preserving yet privacy-protecting through mathematical segmentation.
Solution Approach 2:
The patent introduces an intermediary processing system that operates on segmented data without revealing underlying information. This intermediary layer performs computations and analyses on the shared data while preventing direct access to original personal information, thus enabling data utility for cross-sectional analysis while acting as a mediator that protects privacy by design.
2Object-affected harmful factors
If secret sharing is used to maintain data confidentiality, then privacy protection is improved, but data accessibility and usability for analysis deteriorate
Solution Approach 1:
The patent performs preliminary segmentation of data into shares before distribution to multiple servers. This preliminary action ensures that data is already in a confidential state when accessed by any server, eliminating the need for complex access control mechanisms during analysis operations. The segmentation is done in advance, making subsequent data operations simpler while maintaining confidentiality.
Solution Approach 2:
The patent changes the parameter of data representation from original form to segmented mathematical shares. This parameter transformation allows data to maintain its confidentiality properties while becoming accessible for computational operations. The segmented shares can be manipulated mathematically to perform analyses without ever reconstructing the original personal information, thus improving accessibility while preserving confidentiality.
3Object-affected harmful factors
If anonymization is performed on segmented data, then privacy protection is improved, but data processing complexity increases
Solution Approach 1:
The patent merges the segmentation process with the anonymization process into a unified workflow. Instead of separately segmenting data and then anonymizing it, the system performs both operations in an integrated manner where the segmentation itself contributes to anonymization. This merging reduces processing complexity by eliminating redundant steps while maintaining both anonymity and data utility.
Data Source
AI summary
A processing system distributes registration data that a registrar device has to a plurality of servers in a state of being segmented shares, and stores the registration data in the servers. Each of the servers includes first processing circuitry configured to perform anonymization on the shares, and provide anonymization-processed information on which anonymization is performed in the state of being shares.