Amazon Fraud Detector

Learn about Fraud Detector and pricing model

Amazon Fraud Detector is a fully managed service enabling customers to identify potentially fraudulent activities and catch more online fraud faster. Amazon Fraud Detector features are

  • Automated model creation
  • Trigger actions based on rules
  • Real-time fraud prediction API
  • A single interface to review and audit your predictions and detection logic

How it works

Amazon Fraud Detector

Use Cases

Amazon (AWS) Fraud Detector is predominantly utilized in Fraud detection and Prevention applications and has following use cases

  • Detect new account fraud
  • Identify suspicious online payments
  • Prevent trial and loyalty program abuse

Amazon Fraud Detector Pricing

With Amazon Fraud Detector, you pay only for what you use, and there are no minimum fees or upfront commitments. You are charged based on the compute hours used to train and host your models and per the quantity of fraud predictions you make.

For example, a bank is protecting the customers from high risk purchases. Lets say this model requires 10 GB of training data and trains the model once per month, with each training consuming 24 hours of time. On top of this, You deploy this model for the entire month and use it to generate 5000 fraud predictions per day. Following is the monthly calculator for using Amazon Fraud Detector

Data processing and storage charge = 10 GB x $0.10 per GB = $1

Training charge = 24 compute hours x 1 trainings x $0.39 per compute hour = $9.36

Hosting charge = 30 days x 24 hours x 1 model x $0.06 per compute hour = $43.20

Fraud prediction charge (real-time and batch) = 5,000 predictions / day x 30 days x $0.03 per Online Fraud Insights prediction = $4500

Total cost = $1 + $9.36 + $43.20 + $4500 = $4553.56

More details to be found in –

For more information about other AWS Machine Learning , AI services, check out our post –