AWS releases Amazon Forecast
Developers can now build applications with the same machine learning technology used by Amazon.com
23 August 2019 | 0
Amazon Web Services (AWS) have announced that its fully managed forecasting service, Amazon Forecast, is now generally available.
Forecast allows developers to access the same machine learning technology that Amazon.com uses to predict demand of millions of products. It produces private, custom models to help developers make more accurate predictions that, according to Amazon, are up to 50% more accurate than traditional methods.
The service provisions all necessary infrastructure; processes data; and builds custom machine learning models hosted on AWS.
Developers can tap into Amazon’s extensive forecasting experience now that the service is generally available. Experience in machine learning is not essential to using the service. Customers can achieve accuracy levels that used to take months of engineering in just a few hours, the company stated.
With machine learning, Forecast automatically discovers how variables such as seasonality and store locations affect each other. Such relationships can be hard to spot using traditional forecasting methods. Machine learning recognises patterns so Forecast can set up a data pipeline, ingest data, provide accuracy metrics and perform forecasts.
“We can’t wait to see how our customers use the service to reduce operating expenses and inefficiencies, ensure higher resource and product availability, deliver products faster, and lower costs to delight their customers,” said Swami Sivasubramanian, vice president, Amazon Machine Learning.
Currently, Ireland is the only European country where Forecast is generally available.