How to Build a Real-Time ML Pipeline for Fraud Prediction

Learn how to construct a real-time machine learning pipeline for fraud prediction and prevention in this 47-minute conference talk from the Toronto Machine Learning Series. Explore the challenges of online feature engineering and real-time data ingestion, and discover strategies to overcome them. Gain insights on building an ML pipeline that enables swift ingestion and analysis of real-time data, allowing for immediate action to prevent fraud. Understand the benefits of using a feature store for generating online and offline ML feature calculations in both development and production environments. Delve into techniques for monitoring real-time AI applications in production to detect and mitigate drift, ensuring continued model accuracy despite changing market conditions. Presented by Adi Hirschtein, VP Product at Iguazio, who brings 20 years of experience in driving innovation in technology companies.