Home Cloud Amazon Web Services Amazon Fraud Detector - Explained

Amazon Fraud Detector – Explained


What is Amazon Fraud Detector?

Amazon Fraud Detector is a fully managed service that makes it easy to identify fraudulent online activities such as online payment fraud and fake accounts. Amazon Fraud Detector takes up user’s data, machine learning (ML), and over 20 years of fraud detection expertise to automatically detect fraudulent online activity so scams can be detected at a faster pace.

Amazon Fraud Detector allows the creation of fraud-detection models. ML experience is not a pre-requisite to start using Amazon Fraud Detector; the only requirement is the user providing the company’s historical fraud data. Amazon Fraud Detector uses this data to automatically train, test, and deploy a customized fraud detection model. During this process, serial models that have learned fraud patterns from the user’s account and Amazon’s fraud expertise are used to increase the mechanism’s effectiveness. The customer can use the sample and call the Amazon Fraud Detection “GetEventPrediction” API with metadata about an online event to easily get a result based on a fraud prediction score and their configuration.

Benefits of AWS Fraud Detector

Prevent and detect online fraud – Amazon Fraud Detector is a fraud detection solution that provides everything needed to create, deploy, and manage fraud detection models. With just a few clicks, fraud analysts can improve sample detection with business rules that help control sample behaviour, and then start making predictions into APIs that are ready to produce results.

Fraud detection in minutes – Amazon Fraud Detector automates the complicated steps of creating ML models for fraud detection. Everything from data verification to sample deployment does not require ML or coding experience. Therefore, ML-based fraud detection models can be deployed in minutes instead of months.

Customized for your personal business needs – Amazon Fraud Detector creates the only ML models that combine the fraud patterns learned from user’s business history data with 20+ years of insights from fraud detection on Amazon. The result is a unique, personalized fraud detection model that is optimized for each specific business situation. This approach provides customers with the most accurate fraud detection, coupled with the lowest false positive rate.

It detects online Frauds such as –

New account – It helps users precisely distinguish between legit and high-risk customer account records, so they can selectively introduce additional steps or checks based on risk. For example, a user can set the client account registration workflow so that only account records with high-risk characteristics require additional email and phone verification steps.

Farewell to the guest – Detect potential fraudsters among customers without transaction histories. With Amazon Fraud Detector, a user can send as little as two pieces from a guest update order (e.g., email, IP address) to assess its potential fraud risk, so they can decide whether to accept it, review it, or collect more customer details.

Online payment – Reduce online payment fraud by flagging suspicious online payment transactions before processing payments and executing orders. With Amazon Fraud Detector, a user can set your checkout flow to evaluate new orders and review suspicious items before processing payments.

Online service and loyalty program abuse – It identifies accounts that may abuse online services such as loyalty programs or ‘try before you buy’ programs that allow customers to try the good before paying for them. With Amazon Fraud Detector, online businesses can help customers assess the risk of misusing plans, stealing goods, or misappropriating revenue, by controlling their risks via applying appropriate limits on the value of goods or services provided by businesses.

Akarshan Narang
Akarshan Narang
Covering the world of Cloud at CMI.


Cloud Management