Microsoft Preps Preview Of Cloud-Based Machine Learning Service For Developers


Lots of tech vendors are talking about how machine learning technology is transforming business processes, but Microsoft wants to be one of the first to put this power into the hands of developers.

Next month, Microsoft will release a preview of Azure Machine Learning, a public cloud-based service that lets developers embed predictive analytics into their applications, Joseph Sirosh, corporate vice president of machine learning at Microsoft, said in a blog post Monday.

Azure Machine Learning includes the "powerful algorithms" Microsoft has built for Xbox and Bing, as well as "new analytics tools," Sirosh said in the blog post.

[Related: VMware Reveals Pricing For NSX Software-Defined Networking Tech, Says It's Ready For Channel Rollout]

The main draw is that Microsoft is handling all aspects of running the Azure Machine Learning service, which means developers can get started right away without having to go through the hassle and expense of developing their own machine learning technology in-house.

"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," Sirosh said in the blog post.

As Sirosh notes, machine learning technology gathers and analyzes historical data to predict future results. Online retailers use it today to recommend products to shoppers, and it's also widely used in battling credit card fraud and analyzing traffic patterns.  

Microsoft wants to get machine learning technology into the hands of more enterprises in order to push it into more complex areas, such as predicting disease susceptibility and preventing crime, according to Sirosh.

Chris Hertz, CEO of New Signature, a Washington, D.C.-based Microsoft partner, plans to use the Azure Machine Learning service internally, and also to sell it as a service to his customers.

Hertz told CRN he's always asking his finance, sales and operations teams to come up with predictive intelligence based on New Signature's large stores of historical data.

But this isn't easy to do, and it's also time-consuming, which is why Hertz is interested in buying this technology as a cloud service from Microsoft.

"As a CEO, if I can see farther into the future and with more accuracy, then I can make better and more accurate decisions," Hertz said.  

As a partner, Azure Machine Learning will open a new front for New Signature to deliver services, said Hertz.

Jason Sauers, founder and director of connected systems at Phidiax, a Denver-based Microsoft partner, said Azure Machine Learning is a solid addition to the analytics portion of Microsoft's business intelligence portfolio.

One of the Azure Machine Learning service's big strengths, Sauers said, is that it can analyze complementary sources of data to come up with more focused predictive analytics.

"This could become incredibly valuable, especially to smaller companies who are looking for operations efficiencies but have a limited set of data to learn from," Sauers said.

Microsoft under CEO Satya Nadella is focused on using analytics to improve internal operations and also is pushing hard to establish itself as a big data powerhouse.

By putting a cloud-based machine learning service into the hands of developers, Microsoft is hoping that its cloud will become a place for customers to push the envelope with groundbreaking new types of apps.

PUBLISHED JUNE 16, 2014