Skip to main content

Amazon Personalize

What is it​

A fully managed machine learning service that uses your data to generate item recommendations for your users.

What it's for​

Quickly create and deploy a personalized recommendation engine for your customers, without requiring machine learning expertise.

Use cases​

  • Product recommendations for e-commerce (e.g., "customers who bought X also bought Y").
  • Content recommendations for media platforms (e.g., movies, articles, music).
  • Personalization of user experiences in web and mobile applications.
  • Suggestion of related or complementary items.
  • Generation of trending or popular items lists.

Key points​

  • Fully managed: AWS handles the infrastructure, training, and deployment of ML models.
  • Custom models: Trains ML models using your own user interaction data and item data.
  • Advanced algorithms: Uses cutting-edge ML algorithms, including those used on Amazon.com.
  • Real-time: Can generate recommendations in real-time.
  • No ML expertise required: No prior machine learning knowledge needed to use the service.
  • Integration: Integrates with Amazon S3 for data input and can be accessed via APIs.

Comparison with internally developed recommendation systems:​

  • Amazon Personalize: Significantly reduces the time and effort required to build and maintain a recommendation system, eliminating the complexity of managing ML infrastructure, training models, and deploying them. Allows companies to focus on customer experience personalization rather than ML engineering.
  • Internally developed recommendation systems: Require a team of data scientists and ML engineers, dedicated infrastructure, and significant investment of time and resources to build, train, and maintain models. May offer greater control and customization, but at a much higher cost and complexity.