Amazon is rolling out Amazon Personalize to customers of its Amazon Web Services cloud computing platform. The API-based tool for app developers lets them power personalization features such as product recommendations, search results and direct marketing.
The tool leverages machine learning modeling that Amazon has used in its retail business to drive personalization, but doesn’t require developers to have that particular experience. Amazon stressed that the customer data is kept private at all times.
“We are excited to share with AWS customers the expertise we’ve developed during two decades of using machine learning to deliver great experiences on Amazon.com,” said Swami Sivasubramanian, vice president of machine learning for AWS, Inc. “Customer have been asking for Amazon Personalize and we are eager to see how they implement these services to delight their own end users.”
Amazon Personalize deploys custom, private machine learning models. It provides the necessary structure and manages the entire machine learning pipeline, including processing the data, identifying features, selecting algorithms and training, optimizing and hosting the results.
It’s available initially in Ohio, Northern Virginia, Oregon, Tokyo, Singapore, with broader rollout to follow.
Tech Crunch reported Amazon Personalize’s pricing model charges five cents per gigabyte of data uploaded to Amazon Personalize and 24 cents per training hour used to train a custom model with their data. Real-time recommendation requests are priced on how many are uploaded, with discounts for larger orders.