Coursera Spring Sale
40% Off Coursera Plus Annual!
Grab it
Learn how to generate high-quality synthetic transaction data that maintains the statistical properties of real data while protecting customer privacy in this conference talk from MLOps World. Discover four leading generative approaches—GANs, TVAEs, TabularARGNs, and GPT-based methods—and understand how each can be applied to create secure, realistic datasets for AI applications. Explore the challenges of working with financial and transactional datasets, including mixed data types, rare events like fraud detection, and complex feature dependencies. Follow a practical case study demonstrating data cleaning, encoding, and modeling techniques to train fraud detection systems that remain accurate while ensuring privacy compliance. Master the trade-offs between data utility and privacy protection, learn evaluation methods for models trained on synthetic data, and examine real-world applications across finance, healthcare, and IoT sectors. Gain insights into balancing privacy, utility, and innovation using synthetic data to unlock valuable insights without exposing sensitive customer information.
