Document history - Amazon Fraud Detector

Document history

The following table describes important changes in Amazon Fraud Detector User Guide. We also update the Amazon Fraud Detector User Guide frequently to address the feedback that you send us.

Change Description Date

Create event dataset

Use the guidance provided in Create event dataset to prepare and gather data for training your model.

November 22, 2021

Prediction explanations

Use Prediction explanations in the Amazon Fraud Detector console to get insight into how each event variable impacted your model's fraud prediction scores.

November 10, 2021


Use information in Troubleshoot training data issues to help diagnose and resolve issues you might see in Amazon Fraud Detector console when you train your model.

October 11, 2021

Transaction fraud insights model

Use Transaction fraud insights (TFI) model to detect online or card-not-present transaction fraud.

October 11, 2021

Stored events

You can now store your event data in Amazon Fraud Detector and use the stored data to later train your models. By storing event data in Amazon Fraud Detector, you can train models that use auto-computed variables to improve performance, simplify model retraining, and update fraud labels to close the machine learning feedback loop.

October 11, 2021

Model variable importance

Use model variable importance to gain insight into what is driving performance of your model up or down and which of your model variables contribute the most. And then tweak your model to improve overall performance.

July 9, 2021

Integration with AWS CloudFormation

With this integration, you can now use AWS CloudFormation to manage your Amazon Fraud Detector resources.

May 10, 2021

Batch predictions

Use batch predictions to get predictions for a set of events that do not require real-time scoring.

March 31, 2021

Chapter rework

Rework of Get started and other sections

July 17, 2020

Initial release

Initial release

December 2, 2019