Evaluating your approach to machine learning fraud detection
AWS offers several services that can help with anti-fraud workflows and fraud detection—and avoid the
roadblocks of a do-it-yourself approach:
• Some very specialized fraud detection use cases can be addressed by training your own machine
learning models. While this can be daunting, you can use Amazon SageMaker to simplify and
accelerate the process. Amazon SageMaker is a comprehensive machine learning service that
provides built-in algorithms and pretrained models available through the AWS Marketplace.
• If you do not have access to machine learning expertise—or you prefer to dedicate your machine
learning experts to other areas of your business—you can use Amazon Fraud Detector and Amazon
Rekognition, two AI services that can be integrated into your business applications using an API.
The pre-built machine learning-model templates in Amazon Fraud Detector were developed from 20 years
of experience stopping bad actors from attempting to defraud AWS and Amazon.com. Amazon Fraud
Detector provides templates that you can use to easily create machine learning models that can identify
up to 80 percent more potential bad actors than traditional methods without writing any code. This fully
managed AI and machine learning service provides everything needed to create, deploy, manage, and
scale fraud detection and can be utilized by developers, builders, data scientists, and even business users
with any level of skill in machine learning.
In the following sections, we'll explore how AWS machine learning helps you detect fraud across several
use cases. And we'll look at how specific organizations around the world and across a range of industries
are successfully leveraging AWS machine learning for fraud detection today.
4
