Automate mortgage document fraud detection using an ML model and business-defined rules with Amazon Fraud Detector: Part 3

This article covers how to automate mortgage document fraud detection using Amazon Fraud Detector, an AWS service for building customized fraud detection models. It presents the third component of a mortgage document fraud detection pipeline following two previous posts on automating document validation and building a computer vision model for tampering detection.

Specifically, the article covers:

• Prerequisites for using Amazon Fraud Detector
• Uploading historical data to Amazon S3
• Selecting options and training a fraud detection model
• Creating, reviewing performance, and deploying the model
• Creating a detector and adding business-defined rules
• Deploying the API to make fraud predictions
• Integrating with Amazon SageMaker models (optional)
• Cleaning up resources to avoid future charges