Federated Mining of Anonymized Data Profiles
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Network service providers face challenges in utilizing user profile information for purposes beyond their initial intent due to privacy concerns and regulatory compliance, with existing solutions failing to provide users with control over data usage, access, and duration.
Innovation Solution
A privacy services system that generates anonymized profile identifiers, allowing users to control their data access and usage through permissions management, ensuring secure and anonymous data handling across various network services.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If network service providers utilize user profile information for additional purposes, then productivity and business value are improved, but user privacy and regulatory compliance deteriorate
Solution Approach 1:
The patent introduces a privacy-enhancing technology (PET) system as an intermediary layer between data holders and data users. This mediator enables federated learning by allowing models to be trained across multiple decentralized datasets without exposing raw user data, thus improving data utilization while protecting privacy through cryptographic techniques and secure multi-party computation
Solution Approach 2:
The patent extracts only the necessary computational patterns and statistical insights from user data through federated learning, separating the value extraction process from the raw data itself. This allows providers to mine useful information without directly accessing or exposing sensitive user profile information, resolving the contradiction between data utilization and privacy protection
2Adaptability or versatility
If user profile information is shared with third parties, then business versatility is improved, but user control and permission management deteriorate
Solution Approach 1:
The patent implements dynamic permission management where users can adjust their data sharing preferences, access controls, and permission levels at any time through a user-friendly interface. The system adapts to user choices by dynamically configuring federated learning participation, data access policies, and permission grants, making complex permission management simple and flexible for users
Solution Approach 2:
The patent enables users to self-manage their data permissions and privacy settings through an intuitive interface without requiring technical expertise. Users can independently control which data elements are shared, with whom, and under what conditions, transforming complex permission management into a simple self-service operation
3Quantity of substance
If anonymized data mining is performed across multiple providers, then data quantity and analytical value are improved, but system complexity and security requirements deteriorate
Solution Approach 1:
The patent creates a universal federated learning framework that can operate across multiple network service providers with different data types, formats, and security requirements. The system provides multi-functional capabilities including distributed model training, secure data aggregation, permission management, and result distribution, enabling large-scale collaborative data mining without proportionally increasing system complexity through standardized protocols and modular architecture
Data Source
Figure 1A
Figure 1B
Figure 2
AI summary
Disclosed herein are system, apparatus, article of manufacture, method, and/or computer program product embodiments for efficient and secure data filtering of non-permitted data. An embodiment operates by receiving a federated search request from a profile consumer system identified by a profile consumer identifier, the federated search request including the at least one profile search criteria. The embodiment also operates by transmitting a profile search request to at least one permitted use gatekeeper application, the profile search request including the profile search criteria and receiving a profile search response from the at least one permitted use gatekeeper application, the profile search response including a search result list of users. The embodiment further operates by transmitting a federated search response to the profile consumer system, in response to the federated search request, the federated search response including the federated result list of profile consumers profile identifiers.