Synthetic User Data Profiles for Privacy-Preserving Gap Filling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The handling of user data raises concerns related to privacy and security, and existing methods struggle to efficiently synthesize data for missing periods due to complex relationships between parameters, making it difficult to fill in gaps in user data without exposing sensitive information.
Innovation Solution
A method for generating synthetic data samples involves creating profiles based on existing user data, using intra- and inter-profile parameters, and employing a machine learning model to predict values for these parameters, thereby generating synthetic data that can augment training datasets while maintaining privacy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Object-affected harmful factors
If synthetic data is generated to protect privacy, then user privacy and security are improved, but data completeness and accuracy deteriorate due to gaps in synthetic data
Solution Approach 1:
The system performs preliminary actions by generating profiles from available user data before the gap occurs, and pre-calculates intra-profile parameters that can be carried forward. This allows the synthetic data generation to start with pre-processed information, reducing the impact of data gaps while maintaining privacy protection.
Solution Approach 2:
The patent introduces profiles as an intermediary representation between raw user data and synthetic data. These profiles contain aggregated statistics and patterns that mediate the transformation process, allowing synthetic data generation without directly exposing sensitive individual records, thus addressing both privacy and completeness concerns.
2Measurement precision
If complex relationships between parameters are modeled accurately, then data synthesis quality is improved, but computational complexity increases
Solution Approach 1:
The patent segments the complex parameter relationships into two distinct types: intra-profile parameters (within the same profile) and inter-profile parameters (across different profiles). This segmentation allows different computational strategies to be applied to each type, simplifying the overall computation while maintaining accuracy in modeling parameter relationships.
Solution Approach 2:
The system applies partial action by focusing computational resources on generating only the necessary synthetic data for missing periods, rather than regenerating all data. It uses pre-computed profile statistics and applies them selectively to fill gaps, reducing computational complexity while maintaining synthesis quality.
Data Source
AI summary
Systems and methods for generating synthetic data samples. In some aspects, a system accesses first and second user data samples corresponding to a first and second period of time, respectively, and generates a first and second profile based on the user data samples, wherein the profiles comprise parameters representing metadata associated with corresponding user data samples. The system generates a new profile corresponding to an intermediary period of time between a first and second period of time, wherein the new profile comprises intra-profile and inter-profile parameters. The system determines (1) a value for each intra-profile parameter based on values of intra-profile parameters of the first and second profile and (2) the value for each inter-profile parameter based on a predicted value output of a model trained on a plurality of data profiles over time and generates synthetic data samples based on values of the inter-profile and intra-profile parameters.