Accelerate your business with AWS AI Services
summary
AWS offers pre-trained models you can fine-tune: an 90-minute customisation example yields an AUC of 0.94 using Amazon Fraud Detector, which produces a 0–1000 fraud insight score (example score 811) and shows a threshold tradeoff like 500 → 87% frauds caught with ~13% legitimate false positives. Compute option: P4d (8 A100 GPUs) at about NZ$54/hr. Use Amazon Rekognition for facial comparison in enhanced verification and Amazon Textract to extract structured fields from IDs, paystubs and receipts.
AI-written summary · about
chapters (7)
- 0:00Introduction to AWS AI
- 0:58The AWS AI/ML stack
- 2:31Overview of AI services
- 4:40Amazon fraud detector
- 15:32Amazon rekognition
- 19:31Amazon textract
- 21:47Learning resources
related talks
42mHow to Build Smart Applications Without a Machine Learning Background
39mFraud Detection: Using ML to Identify and Manage Fraudulent Activities
31mCatch more potential online fraud faster with Amazon Fraud Detector
01h 06mAmazon Machine Learning to Predict Responses
38mUsing Artificial Intelligence to Verify Identity
25mTransform your business with data-driven insights
29mSolve common business problems with AI and ML
15mAI as a Service: Quick and Scalable Machine Learning POCs with No-Code/Low-Code Tools