Build a no-code ML workflow with Snowflake, Amazon SageMaker Canvas and Amazon Quick – Part 2: Data preparation and model building with Amazon SageMaker Canvas

This article demonstrates data preparation and fraud detection model building using Amazon SageMaker Canvas with Snowflake integration, requiring no machine learning expertise.

• Connect Amazon SageMaker Canvas directly to Snowflake data sources for real-time data access
• Use Data Wrangler to visually transform data, join multiple sources, and engineer features like outlier thresholds
• Create custom formulas to flag unusual transactions based on card and merchant spending patterns
• Remove sensitive columns (card numbers, merchant names) to maintain data governance
• Run quality analysis reports to identify data issues and preview model accuracy before training
• Train a 2-category fraud detection model using XGBoost ensemble method in 15-30 minutes
• Generate predictions on unseen datasets and send results to Amazon QuickSight for visualization

The no-code workflow enables business analysts to build production-ready ML models while maintaining enterprise security and governance standards.