Overview
Aurora integrates directly with AWS Machine Learning services, allowing you to run ML-based predictions directly from SQL queries without building custom integration code. This provides a simple, optimized, and secure way to enhance applications with AI capabilities.
Supported AWS Services
- Amazon SageMaker – Use any trained ML model for real-time predictions.
- Amazon Comprehend – Perform sentiment analysis, entity recognition, and text classification.
Key Features
- No ML expertise required – Developers can call ML models from standard SQL.
- Secure integration – Data is transferred securely between Aurora and ML services.
- Low-latency predictions – Results are returned as part of the query response.
- Optimized for performance – Aurora manages connections to ML services efficiently.
Example Use Cases
- Fraud detection – Evaluate transactions in real time.
- Ad targeting – Deliver personalized ads based on prediction scores.
- Sentiment analysis – Assess customer feedback directly in SQL queries.
- Product recommendations – Suggest items based on purchase history.
How It Works
- Application sends a SQL query (e.g.,
SELECT ml_predict(...)).
- Aurora sends the relevant data (e.g., user history) to the ML service.
- AWS ML Service processes the request and returns predictions.
- Aurora includes predictions in the SQL query result set for the application.