Announcing Smart Data Validation for Amazon Fraud Detector

Today, Amazon Fraud Detector (AFD) announced the launch of Smart Data Validation, a new feature which helps customers streamline their data preparation process for a successful batch import. Smart Data Validation helps customers save time and resources by validating data at the time of import. This new feature reduces the time to validate and fix customer’s data from days to minutes.

Before this launch, data preparation in AFD could be a complex and time-consuming process for customers. It required manual effort to identify, clean, and validate the data before importing it to AFD for training a ML fraud model. It also required customers to follow carefully after data guidelines, as each of Amazon Fraud Detector fraud models has unique requirements on the quality and format of the data. Previously, this process resulted with customers making a few attempts before successfully importing data into AFD.

Smart Data Validation now automates the process of data validation during data import. It identifies abnormalities in your data and provides with a simple and effective data validation report to help ensure that each data point is formatted properly and ready for training a ML fraud model. It also helps clear up ambiguities around data requirements and formatting. Starting today, customers can ensure their data is accurate and complete in a fraction of the time.

This feature is available in all regions where Amazon Fraud Detector is available: US East (Ohio), US East (N. Virginia), US West (Oregon), Europe (Ireland), Asia Pacific (Singapore) and Asia Pacific (Sydney). For additional details, see our documentation and pricing page.