Amazon launches Personalize, a fully managed AI-powered recommendation service

Amazon today announced the general availability of Amazon Personalize, an AWS service that facilitates the development of websites, mobile apps, and content management and email marketing systems that suggest products, provide tailored search results, and customize funnels on the fly.

It’s available in select AWS regions to start, including US East (Ohio), US East (North Virginia), US West (Oregon), Asia Pacific (Tokyo), Asia Pacific (Singapore), and EU (Ireland). Additional locations are on the way.

“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 AWS’ VP of machine learning Swami Sivasubramanian. “Customers have been asking for Amazon Personalize, and we are eager to see how they implement these services to delight their own end users. And the best part is that these artificial intelligence services, like Amazon Personalize, do not require any machine learning experience to immediately train, tune, and deploy models to meet their business demands.”

Personalize, which was announced last year at Amazon’s re:Invent conference, is a fully managed service that trains, tunes, and deploys custom machine learning models in the cloud by provisioning the necessary infrastructure and managing things like data processing, feature extraction, algorithm training and optimization, and hosting. Customers provide an activity stream from their apps and websites — e.g., clicks, page views, signups, and purchases — in addition to an inventory of the items they want to recommend (such as articles, products, videos, or music) and optional demographic information (like age or geographic location). They receive results via an API and only pay for what they use.

Above: A diagram illustrating the Amazon Personalize process.

Image Credit: Amazon

Amazon charges five cents per GB of data uploaded to Personalize and 24 cents per training hour used to train a custom model. Real-time recommendation requests are priced based on how many requests are uploaded, with discounts for larger orders.

Domino’s, Yamaha, Subway, Bollywood on-demand video host Spuul, and wedding company Zola are already using Personalize to highlight musical instruments in store catalogs, deliver ingredient and flavor recommendations, and devise individualized style combinations.

“At Subway, guest experience matters. Using Amazon Personalize, we can quickly deliver personalized recommendations for our endless varieties of ingredients and flavors to fit the unique lifestyles of our busy guests,” said Subway executive Neville Hamilton. “Amazon Personalize lets our team … curate recommendations without requiring machine learning expertise. We are looking forward to continuing to work with Amazon Personalize to provide the best experience to our guests who want to eat fresh. We have already successfully tested using Amazon Personalize to provide recommendations to guests making orders from our app, and are excited to expand into personalized app notifications in the near future.”

Personalize is the latest fully managed AWS service to hit general availability after Textract, which uses machine learning to parse data tables, forms, and whole pages for text and data. In somewhat related news, Amazon in March launched AWS Deep Learning Containers, a library of Docker images preinstalled with popular deep learning frameworks.