MUMBAI, India, May 1 -- Intellectual Property India has published a patent application (202641050184 A) filed by Sr University, Warangal, Telangana, on April 20, for 'method and system for ai-driven synthetic data generation using ctgan.'
Inventor(s) include Dr. Neha Rathod; Dr. Sumit Kumar Singh; and Dr. Iram Nausheen.
The application for the patent was published on May 1, under issue no. 18/2026.
According to the abstract released by the Intellectual Property India: "Method and System for AI-Driven Synthetic Data Generation Using CTGAN The present invention relates to an artificial intelligence (AI)-driven system and method for generating synthetic tabular data using deep learning-based machine learning models, specifically Conditional Generative Adversarial Networks (CTGAN). The system comprises a data acquisition and preprocessing module configured to handle heterogeneous datasets including continuous and categorical variables, a generative learning module implementing a generator-discriminator architecture, and a validation and privacy assessment module. The generator produces synthetic data samples conditioned on selected attributes, while the discriminator evaluates authenticity through adversarial training. The system employs mode-specific normalization and conditional sampling to accurately capture complex statistical distributions, inter-feature correlations, and imbalanced class characteristics. Further, the system integrates statistical similarity evaluation, machine learning efficacy testing, and privacy risk assessment to ensure fidelity and prevent data memorization. The disclosed invention enables scalable, privacy-preserving synthetic data generation suitable for applications including healthcare, finance, and data analytics, thereby improving data accessibility and model performance."
Disclaimer: Curated by HT Syndication.
