Cross-Entity Data Analysis Using MPC Secret Sharing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems face challenges in analyzing online activity data across multiple entities without revealing sensitive user information, leading to potential privacy breaches.
Innovation Solution
A method involving multi-party computation (MPC) and secret sharing techniques, where data is distributed among multiple MPC devices to perform analysis while maintaining privacy, using secret shares and random noise to protect user data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If data is centralized for analysis, then analysis accuracy is improved, but user privacy is compromised
Solution Approach 1:
The patent segments user data into secret shares distributed across multiple MPC devices. Each device holds only a portion of the data (secret shares), preventing any single device from accessing complete user information. The analysis is performed on segmented data across distributed devices, maintaining both accuracy through collective computation and privacy through data segmentation.
Solution Approach 2:
The patent introduces random noise as an intermediary element in the computation process. This noise is added to the data during multi-party computation to obscure individual user information while preserving statistical properties needed for accurate analysis. The noise acts as a mediator that enables analysis without directly exposing sensitive user data.
2Object-affected harmful factors
If secret sharing is implemented across multiple devices, then user privacy is protected, but processing speed decreases
Solution Approach 1:
By segmenting data into secret shares and distributing them across multiple devices, the system enables parallel processing of different data portions. Each MPC device can independently process its assigned secret shares simultaneously, reducing overall processing time compared to sequential processing on a single device while maintaining privacy through distributed storage.
Solution Approach 2:
The patent performs preliminary actions by pre-processing and distributing secret shares to multiple MPC devices before the actual analysis computation. This preparation phase enables the devices to be ready for parallel computation, significantly accelerating the main analysis process while the privacy protection through secret sharing is already established.
3Object-affected harmful factors
If random noise is added to protect privacy, then privacy is improved, but result accuracy deteriorates
Solution Approach 1:
The patent carefully controls the parameters of the added random noise, specifically its magnitude and distribution characteristics. By adjusting noise parameters to optimal levels, the system achieves sufficient privacy protection while minimizing the degradation of result accuracy. The noise parameters are tuned to provide the minimum necessary obfuscation for privacy while preserving statistical properties for accurate analysis.
Solution Approach 2:
The system employs feedback mechanisms to evaluate the impact of added noise on analysis results and adjust noise parameters accordingly. By monitoring result quality and privacy protection effectiveness, the system can iteratively optimize noise levels to achieve the best balance between privacy protection and result accuracy, removing excessive noise that degrades accuracy while maintaining sufficient noise for privacy.
Data Source
AI summary
Methods, systems, and apparatus, including computer programs encoded on a computer storage medium. In one aspect, a method includes receiving, from a content distributor, plan data specifying a set of distribution plans that cause distribution of content. Instructions are transmitted to publishers to submit secret shares of a multi-register sketch representing presentations of the content. A notification that the content distributor has requested an analysis of the presentations of the content is sent to a multi-party computing group. A result share of the analysis of the presentation of the content is received from multiple MPC devices in the MPC group. A set of result shares received from the of MPC devices are transmitted to the content distributor.