Training a call center fraud detection model for IVR calls with Amazon SageMaker Canvas

This article discusses how to build a fraud detection model for call center IVR calls using Amazon SageMaker Canvas, a no-code/low-code machine learning service.

• Uses Amazon Connect contact trace record (CTR) dataset for training
• Enriches raw data using Amazon Pinpoint's phone number validation API
• Prepares and cleans data to improve model accuracy
• Builds and trains fraud detection model in SageMaker Canvas
• Explores deployment options including autoscaling and serverless inference

The solution enables call centers to detect potential fraud efficiently without compromising customer experience, leveraging machine learning to automate fraud screening.