Determining your use case - 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.

Determining your use case

Before you use Amazon Personalize, determine your use case to identify what recipe to use to train the model and what data to import. Recipes are Amazon Personalize algorithms that are prepared for specific use cases. After you import data, you tell Amazon Personalize what recipe to use.

Amazon Personalize use cases

Before you get started providing personalized experiences for your users, choose your use case from the following and note its corresponding recipe type.

  • Recommending items for users (USER_PERSONALIZATION recipes)

    To provide recommendations for your users, train your model with a USER_PERSONALIZATION recipe. Recommendations can either be personalized recommendations, such as recommending movies for a user based on interactions, items, and user data, or popular items based on interactions data.

  • Ranking items for a user (PERSONALIZED_RANKING recipes)

    To personalize the order of curated lists or search results for your users, train your model with a PERSONALIZED_RANKING recipe. PERSONALIZED_RANKING recipes create a personalized list by re-ranking a collection of input items based on predicted interest level for a given user. Personalized lists improve the customer experience and increase customer loyalty and engagement.

  • Recommending similar items (RELATED_ITEMS recipes)

    To recommend similar items, such as items frequently bought together or movies that other users have also watched, you should use a RELATED_ITEMS recipe. Recommending similar items can help your customers discover items and can increase user conversion rate.

  • Getting user segments (USER_SEGMENTATION recipes)

    To get segments of users based on item input data, such as users who will most likely interact with items with a certain attribute, you should use a USER_SEGMENTATION recipe. Getting user segments can help you create advanced marketing campaigns that promote different items to different user segments based on the likelihood that they will take an action.

What data to import

If you are providing personalized recommendations for your users, import data that helps Amazon Personalize recommend items that are new or have fewer user interactions. This data includes creation timestamp data and unstructured text metadata about your items, as well as impressions data and contextual metadata from your customers' interactions with items.

For personalized recommendations or ranked items, import categorical data, such as a user's gender or an item's genre, to help Amazon Personalize identify underlying patterns that reveal the most relevant items for your users. For all use cases, you can use categorical metadata to filter recommendations from Amazon Personalize based on a user or item's attributes.

For general information about how Amazon Personalize reads and stores your data, see Datasets and schemas. For more information on the types of data you can import, see Interactions data, Item data, and User data.