Quantized Data Sharing for Multi-Statistic Privacy and Low Distortion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional data sharing tools fail to protect multiple summary statistics properties of shared data, which are often considered business secrets, lacking a generalized framework to ensure privacy in multi-dimensional scenarios.
Innovation Solution
Implementing a platform, language, and cloud agnostic data sharing module that utilizes quantization-based methods to analyze risks and provide provable privacy guarantees by defining privacy and distortion metrics, minimizing distortion while preserving confidentiality through algorithms for Gaussian distributions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional data sharing tools are used, then data sharing is enabled, but multiple summary statistics properties (business secrets) are not protected
Solution Approach 1:
The patent extends the privacy framework from protecting a single summary statistic to protecting multiple summary statistics simultaneously. The generalized framework defines privacy metrics and distortion metrics that work across multiple secrets, making the system universal for various data sharing scenarios involving multiple business secrets.
Solution Approach 2:
The patent introduces new parameters including privacy metrics (measuring how well multiple secrets are protected), distortion metrics (measuring the trade-off between privacy and data utility), and tolerance ranges for each secret. These parameter changes enable the system to quantify and optimize privacy protection across multiple summary statistics.
2Reliability
If privacy protection is strengthened for multiple secrets, then confidentiality is improved, but data utility and accuracy may deteriorate
Solution Approach 1:
The patent introduces tolerance ranges for each secret that can be adjusted to balance privacy protection and data utility. By changing these parameters, users can optimize the trade-off: stricter tolerance ranges provide stronger privacy but reduce utility, while more relaxed ranges improve utility but weaken privacy protection.
Solution Approach 2:
The framework allows selective protection of different secrets with different tolerance ranges, enabling partial privacy protection where some secrets require stronger protection than others. This partial action approach optimizes the balance between privacy and utility by applying different levels of protection where needed.
Data Source
AI summary
Various methods and processes, apparatuses or systems, and media for protecting confidential aggregate dataset information when sharing data are disclosed. A receiver receives confidential dataset from a data owner via a communication interface, the confidential dataset including a multi-dimensional privacy data, and being generated from an original distribution of dataset as released distribution dataset. A processor, operatively connected to the receiver, defines a privacy metric as a probability of an attacker guessing the multi-dimensional privacy data by applying a first data processing algorithm onto the confidential dataset; defines a distortion metric of a data release mechanism as worst-case distance between the original distribution dataset and the released distribution dataset by applying a second data processing algorithm; and implements the data release mechanism that minimizes the distortion metric subject to a constraint on the privacy metric for protecting the confidential aggregate dataset information when sharing data.