Amazon Personalize workflow summary - Amazon Personalize
Services or capabilities described in Amazon Web Services documentation might vary by Region. To see the differences applicable to the China Regions, see Getting Started with Amazon Web Services in China (PDF).

Amazon Personalize workflow summary

The Amazon Personalize workflow is as follows:

  1. Create a dataset group

    A dataset group is a container for Amazon Personalize resources. The type of dataset group you create determines the resources you can create in step 3 of the Amazon Personalize workflow.

    • With a Domain dataset group, you can create recommenders for VIDEO_ON_DEMAND or ECOMMERCE domain use cases. Amazon Personalize manages the configuration, training and updating of these recommenders. If you start with a Domain dataset group, you can still add custom resources.

    • With a Custom dataset group, you can create only custom resources. These including solutions, solution versions, and campaigns. For these resources, you have more control over configurations, updates, and retraining.

  2. Prepare and import data

    You import interaction, item, user, action, and action interaction records into datasets (Amazon Personalize containers for data). You can import records in bulk or individually. When you import bulk data, you can use Amazon SageMaker Data Wrangler to import data from 40+ sources and prepare it for Amazon Personalize. For more information, see Preparing and importing data using Amazon SageMaker Data Wrangler.

    After you import data into an Amazon Personalize dataset, you can analyze it, export it to an Amazon S3 bucket, update it, or delete it by deleting the dataset. For more information, see Managing data.

  3. Create domain recommenders or custom resources

    After you import your data, create domain recommenders (for Domain dataset groups) or custom resources (for Custom dataset group) to train a model on your data. You use these resources to generate recommendations.

  4. Get recommendations

    Use your recommender or custom campaign to get recommendations. With a Custom dataset group, you can also get batch recommendations or user segments.

After you complete the Amazon Personalize workflow the first time, keep your data current, and regularly re-train any custom solutions. This allows your model to learn from your user’s most recent activity and sustains and improves the relevance of recommendations. For more information, see Maintaining recommendation relevance.