Amazon Forecast, a fully managed event forecasting service using machine learning, is now widely available. Predictions about, for example, sales, required infrastructure or power requirements are thus among the possibilities for companies.
Amazon reported this in a press release on Wednesday. Forecast uses machine learning to make predictions in different fields, and also links cloud infrastructure provisioning to it. No AI or machine learning expertise is required to use this service, hence ‘fully managed’. Amazon Forecast automatically creates a data pipeline and processes data to train a model. Accuracy data is then calculated and the service executes predictions.
Machine learning predictions also for non-experts
According to Amazon, the system can build complex machine learning models in a few simple steps. That would normally take a long time. The company also claims that the predictions are 50 percent more accurate than traditional models. On the one hand, this is because there are enormous amounts of historical data available that Forecast can work with. On the other hand, it also includes related, but separate data that is not used or is used less in traditional methods.
Swami Sivasubramanian, Vice President at Amazon Machine Learning, says in the press release: “Over the years, we have developed advanced predictive algorithms that our customers can now use in Amazon Forecast without having to know anything about machine learning. We can’t wait to see how our customers use the service to reduce their operating costs and inefficiencies, ensure higher availability of resources and products, deliver products faster and reduce costs to satisfy their customers.”
This news article was automatically translated from Dutch to give Techzine.eu a head start. All news articles after September 1, 2019 are written in native English and NOT translated. All our background stories are written in native English as well. For more information read our launch article.