Overview
Maintaining Personalized Experiences with Machine Learning helps you build custom Amazon Personalize experiences for your product portfolio, including custom recommendation models at scale. This AWS Solution streamlines and accelerates the development and deployment of your personalization workloads through automation and scheduled updates for resources within Amazon Personalize.
Benefits
Automate the creation of all resources in Amazon Personalize upfront to save on time and costs.
Integrate workflows around Amazon Personalize into your applications.
Technical details
You can automatically deploy this architecture using the implementation guide and the accompanying AWS CloudFormation template.
Step 1
The AWS CloudFormation template deploys an Amazon Simple Storage Service (Amazon S3) bucket used to store personalization data and configuration files.
Step 2
An AWS Lambda function initiated when new or updated personalization configuration is uploaded to the personalization data bucket.
Step 3
An AWS Step Functions workflow manages all of the resources of an Amazon Personalize dataset group (including datasets, schemas, event tracker, filters, solutions, campaigns, and batch inference jobs).
Step 4
Amazon CloudWatch metrics for Amazon Personalize for each new trained solution version are added to help you evaluate the performance of a model over time.
Step 5
An Amazon Simple Notification Service (Amazon SNS) topic and subscription notifies an administrator when the maintenance workflow has completed via email.
Step 6
Amazon DynamoDB tracks the scheduled events configured for Amazon Personalize to fully or partially retrain Amazon Personalize solutions, import or reimport datasets, and perform batch inference jobs.
Step 7
A Step Functions workflow tracks the current running scheduled events, and invoke step functions to perform Amazon Personalize solution maintenance (creating new solution versions, updating campaigns), import updated datasets, and perform batch inference.
Step 8
A set of maintenance Step Functions creates new dataset import jobs on schedule; performs Amazon Personalize solution FULL retraining on schedule (and update associated campaigns); performs Amazon Personalize solution UPDATE retraining on schedule (and update associated campaigns); and creates batch inference jobs.
Step 9
Resource status notification updates are posted to an Amazon EventBridge event bus throughout the Step Functions workflow.
Step 10
A command line interface (CLI) allows you to import and establish schedules for resources that already exist in Amazon Personalize.
Step 1
The AWS CloudFormation template deploys an Amazon Simple Storage Service (Amazon S3) bucket used to store personalization data and configuration files.
Step 2
An AWS Lambda function initiated when new or updated personalization configuration is uploaded to the personalization data bucket.
Step 3
An AWS Step Functions workflow manages all of the resources of an Amazon Personalize dataset group (including datasets, schemas, event tracker, filters, solutions, campaigns, and batch inference jobs).
Step 4
Amazon CloudWatch metrics for Amazon Personalize for each new trained solution version are added to help you evaluate the performance of a model over time.
Step 5
An Amazon Simple Notification Service (Amazon SNS) topic and subscription notifies an administrator when the maintenance workflow has completed via email.
Step 6
Amazon DynamoDB tracks the scheduled events configured for Amazon Personalize to fully or partially retrain Amazon Personalize solutions, import or reimport datasets, and perform batch inference jobs.
Step 7
A Step Functions workflow tracks the current running scheduled events, and invoke step functions to perform Amazon Personalize solution maintenance (creating new solution versions, updating campaigns), import updated datasets, and perform batch inference.
Step 8
A set of maintenance Step Functions creates new dataset import jobs on schedule; performs Amazon Personalize solution FULL retraining on schedule (and update associated campaigns); performs Amazon Personalize solution UPDATE retraining on schedule (and update associated campaigns); and creates batch inference jobs.
Step 9
Resource status notification updates are posted to an Amazon EventBridge event bus throughout the Step Functions workflow.
Step 10
A command line interface (CLI) allows you to import and establish schedules for resources that already exist in Amazon Personalize.
Related content
This video shows you how to streamline and accelerate the development, automation, and deployment of your Amazon Personalize workloads using Maintaining Personalized Experiences with Machine Learning.
- Publish Date