Maintaining recommendation relevance

Relevant recommendations can increase user engagement, click-through rate, and conversion rate for your application as your catalogue grows. To maintain and improve the relevance of Amazon Personalize recommendations for your users, keep your data and custom resources up to date. This allows Amazon Personalize to learn from your user’s most recent behavior and include your newest items in recommendations.

Keeping datasets current

As your catalog grows, update your historical data with bulk or individual data import operations. For more information about importing historical data, see Step 2: Preparing and importing data. For information on how data you import after training a model influences recommendations, see How new data influences real-time recommendations.

For use cases and recipes that provide personalized real-time recommendations, keep your Item interactions dataset up to date with your users' behavior. Do this by recording item interactions with an event tracker and the PutEvents API operation. Amazon Personalize updates recommendations based on your user's most recent activity as they interact with your catalog. For information about real-time personalization, see Real-time personalization. For more information on recording real-time events, see Recording events.

Maintaining domain recommenders

Amazon Personalize automatically retrains the models backing your recommenders every 7 days. This is a full retraining that creates entirely new models based on the entirety of the data in your datasets. If you modify the columns used in training, Amazon Personalize automatically starts a full retraining of the models backing your recommender.

• For Top picks for you and Recommended for you use cases, Amazon Personalize updates your recommender to consider new items for recommendations. Automatic updates are not a full retraining where the model learns from your users' behavior. Instead, automatic updates allow Amazon Personalize to feature your new items in recommendations before the recommender's next full retraining. For information about automatic updates, see Automatic updates.

• If you use the Trending now use case, Amazon Personalize automatically evaluates your interactions data every two hours and identifies trending items. You don't have to wait for your recommender to retrain.

While recommender retraining is in progress, you can still get recommendations from the recommender. Until the retraining completes, the recommender uses the previous configuration and models. To track updates, you can view the timestamp for the latest recommender update on the Recommender details page in the Amazon Personalize console. Or you can view the latestRecommenderUpdate details from the DescribeRecommender operation.

Maintaining custom solutions

By default, all new solutions use automatic training to create a new solution version every 7 days. Training continues until you delete the solution.

When you create a solution, we recommend that you use automatic training to manage solution version creation. This makes maintaining your solution easier. It removes the manual training required for the solution to learn from your more recent data. Without automatic training, you must manually create new solution versions for the solution to learn from your most recent data. For more information about configuring automatic training, see Configuring automatic training.

Your training frequency depends on your business requirements, the recipe that you use, and how frequently you import data. For all recipes, we recommend training at least weekly. With automatic training, this is the default training frequency. If you frequently add new items or actions, you might want to have a higher training frequency, depending on your recipe.

• If you use User-Personalization or Next-Best-Action, the solution automatically updates to consider new items or actions for recommendations. Automatic updates aren't the same as automatic training. An automatic update doesn't create a completely new solution version, and the model doesn't learn from your latest data. To maintain your solution, your training frequency should still be at least weekly. For more formation about automatic updates, including additional guidelines and requirements, see Automatic updates.

• If you use Trending-Now, Amazon Personalize automatically identifies the top trending items in your interactions data over a configurable interval of time. Trending-Now can recommend items added since the last training through bulk or streaming interactions data. Your training frequency should still be at least weekly. For more information, see Trending-Now recipe.

• If you don't use a recipe with automatic updates or the Trending-Now recipe, Amazon Personalize considers new items for recommendations only after the next training. For example, if you use the Similar-Items recipe, and you add new items daily, you must use a daily training frequency for these items to appear in recommendations that same day.

You can automate and schedule retraining and data import tasks with Maintaining Personalized Experiences with Machine Learning, which is an AWS Solutions Implementation that automates the Amazon Personalize workflow, including data import, solution version training, and batch workflows. For more information, see Maintaining Personalized Experiences with Machine Learning.