Microsoft is adding predictive analysis to its repertoire through its Azure cloud computing program. The new service, called Azure Machine Learning, will give businesses access to advanced predictive software that can be adapted to nearly any situation.
Azure ML takes advantage of the predictive modeling techniques Microsoft already uses for its Bing search engine as well as the Xbox as a base for users to create programs to analyze their own data. The service will feature a user friendly visual interface that contains templates for common tasks and icons representing common functions that can be dragged and dropped into the system.
Most corporations that use machine learning techniques do so through in-house data analysis using private servers. Software engineers can spend months working to create a predictive model that can fulfill the needs of that particular business. Microsoft is taking a more modular approach. Azure ML can be built upon to create a predictive model for almost any task. Since it's embedded into the cloud computing service, it also doesn't require organizations to purchase and maintain expensive servers to perform data calculations.
"In mere hours, with Azure ML, customers and partners can build data-driven applications to predict, forecast and change future outcomes - a process that previously took weeks and months," says Microsoft in a statement.
The adaptability of Azure ML makes it a valuable resource for almost any industry. Predictive computing can be used to warn the maintenance department of a company when a part may be near failure. This could prevent planes from being stranded on the runway or manufacturers from being forced to issue recalls. In the confidential testing period for Azure ML, it has already been used to help a major retailer predict future sales. It is also being used at Carnegie Mellon University to analyze the power consumption of the buildings on campus and determine the cause of any variations in energy usage.
Azure ML supports more than 300 packages using the R programming language; allowing users to assemble a model suited to their needs out of existing pieces rather than be forced to build something from scratch. Data already stored on Azure can be analyzed more quickly to create a working method of predictive analysis. However, users can also stream data from their own computers through Azure ML to allow the program to gather enough data to make predictions.
Azure ML has been in use by about 100 private preview customers for nearly a year, and a public preview is scheduled to begin in July. The full release date of the service has not yet been announced.
