Row Conditional-TGAN for generating synthetic relational databases

TL;DR

This paper introduces RC-TGAN, a novel GAN model that effectively generates synthetic relational databases by modeling relationships between tables, including complex multi-level dependencies, improving data quality and relationship preservation.

Contribution

The paper presents RC-TGAN, a new GAN architecture that models and synthesizes relational databases with inter-table relationships, including grandparent-grandchild dependencies.

Findings

Significant improvement in synthetic data quality over benchmarks.

Effective modeling of parent-child and grandparent-grandchild relationships.

Demonstrated robustness across eight real relational databases.

Abstract

Besides reproducing tabular data properties of standalone tables, synthetic relational databases also require modeling the relationships between related tables. In this paper, we propose the Row Conditional-Tabular Generative Adversarial Network (RC-TGAN), a novel generative adversarial network (GAN) model that extends the tabular GAN to support modeling and synthesizing relational databases. The RC-TGAN models relationship information between tables by incorporating conditional data of parent rows into the design of the child table's GAN. We further extend the RC-TGAN to model the influence that grandparent table rows may have on their grandchild rows, in order to prevent the loss of this connection when the rows of the parent table fail to transfer this relationship information. The experimental results, using eight real relational databases, show significant improvements in the quality of the synthesized relational databases when compared to the benchmark system, demonstrating the effectiveness of the RC-TGAN in preserving relationships between tables of the original database.