Generative AI Data Synthesizer for Imbalanced Tabular Banking Datasets
Conditional Tabular GAN (CTGAN) and TVAE engine generating mathematically valid synthetic credit records with preserved correlations.
Project Overview
Empowers data science teams to train models on proprietary banking data without violating customer privacy. CTGAN models mixed continuous and discrete data distributions, producing synthetic datasets that pass Kolmogorov-Smirnov correlation tests.
Empowers data science teams to train models on proprietary banking data without violating customer privacy. CTGAN models mixed continuous and discrete data distributions, producing synthetic datasets that pass Kolmogorov-Smirnov correlation tests.
Conditional Tabular GAN (CTGAN) and TVAE engine generating mathematically valid synthetic credit records with preserved correlations.
Core Project Objectives
Capture continuous analog/digital sensor readings with robust noise filtering and hardware calibration.