Synthetic Relational Data Linking for Preserved Table Relationships
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for generating synthetic data fail to effectively replicate relationships between tables in relational datasets, leading to inefficient and computationally expensive solutions that do not adequately mimic real-world scenarios.
Innovation Solution
Generate extended tables based on a reference relational dataset, incorporating data from parent and child tables, and use methods like statistical or deep learning to create synthetic extended tables independently, then determine links between them based on similarity measures to replicate relationships.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If existing methods for generating synthetic data are used, then individual tables can be synthesized, but relationships between tables are not retained or replicated
Solution Approach 1:
The patent segments the relational dataset into individual tables while preserving relationship information separately. Each table is processed independently to generate synthetic data, then relationships are reconstructed by matching foreign keys with primary keys across synthetic tables, maintaining referential integrity without treating the entire dataset as one complex unit
Solution Approach 2:
The patent performs preliminary actions by first generating synthetic data for each table independently before establishing relationships between them. This allows each table to be synthesized with its own statistical properties first, then relationships are added in a subsequent step by matching keys, simplifying the overall process
2Manufacturing precision
If large real-world datasets are used for testing, then comprehensive testing scenarios are available, but storage and processing resources are substantially required
Solution Approach 1:
The patent creates synthetic copies of real-world datasets that replicate statistical properties, data distributions, and relationship structures without copying the actual data values. This produces smaller datasets that maintain testing fidelity while reducing storage requirements and processing demands
3Productivity
If synthetic data is generated to reduce storage requirements, then resource efficiency improves, but relationships between tables may be lost
Solution Approach 1:
The patent uses feedback mechanisms by generating synthetic tables with foreign keys that reference primary keys in other synthetic tables. The system verifies that relationships are properly established by checking referential integrity, and adjusts the generation process to ensure relationship structures are preserved in the reduced-size synthetic dataset
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Methods and apparatus for generating synthetic data based on a reference relational dataset comprising a plurality of tables and a plurality of relationships between the tables. The methods include generating extended tables for each table in the reference relational dataset and generating synthetic extended tables for each extended table. Measures of similarity are then determined between data records in child and parent synthetic extended tables and links between data records determined based on the measures measure of similarity. Generated synthetic data includes data from the synthetic extended tables and for child tables, a foreign key field populated based on the determined links between data records child and parent synthetic extended tables.