Adjusted R² measures the goodness of fit of a regression model. Higher the R², better is the model.

Overall it was an inspiring week with many deep engaging conversations. Visitors to the Microsoft area were excited to see customer success stories from us and our partners and learned about the details of the Azure services that were used to build them. Hannover Messe this year featured many smart technologies and digital transformation breakthroughs tailored to the world of manufacturing and we look forward to being back next year with even more exciting updates to share.

The news and information ecosystem is in the midst of change -- again. Mobile-first consumption is on the rise, smart homes are becoming mainstream and connected cars will soon take over the roads of major cities around the world. Smart devices will require "smart content." It's only a matter of time before artificial intelligence becomes the backbone of the media industry of the future. Today, most people find information via search or social.

This story was delivered to BI Intelligence "E-Commerce Briefing" subscribers. To learn more and subscribe, please click here. The online marketplace has acquired AI company Expertmaker to enhance the way products display on eBay's pages, according to Internet Retailer. This purchase is part of eBay's structured data push for sellers, which utilizes eBay's standard way of categorizing and displaying products for sale on its marketplace. The use of structured data is not mandatory, but eBay has been strongly encouraging sellers to adopt it. As of the first quarter of 2016, 60% of listed items on eBay used the structured data rules, according to the company's earnings report.

Venue Sponsor: WeWork - "WeWork is the platform for creators. We provide the workspace, community, and services you need to make a life, not just a living.

Celaton's inSTREAM software streamlines the handling of unstructured unpredictable (and structured) content such as correspondence, claims, complaints and invoices that organizations receive by email, social media, fax and paper. This minimizes the need for human intervention and ensures that only accurate, relevant and structured data enters business systems. Unique to inSTREAM is its ability to learn through the natural consequence of processing information and collaborating with people. Capgemini's extensive knowledge and experience in business process services will also enable Celaton to accelerate and improve inSTREAM's capabilities. The cooperation will enable Capgemini to increase efficiency, shorten turnaround times and enhance quality in areas where incoming documents and queries need to be processed, improving overall customer satisfaction.

When that quadrotor fencing video showed up everywhere last month, we asked Ross Allen, the Stanford PhD candidate (and fencer) responsible for the research, if he'd be willing to talk to us about it. He said sure, except his thesis defense was that Friday, so would we mind waiting a bit? It's been a bit, and after a successful defense, Dr. Allen is somehow not sick and tired of robots and answered a bunch of our questions about quadrotors, swords, and why mixing them is such a great idea. Here's the video that you (and a couple hundred thousand other people) probably saw a few weeks ago: The swordplay is cool, but even cooler is the fact that this is the first demonstration of truly "real-time kinodynamic planning" on a quadrotor system navigating an obstructed environment. Or at least, that's what this recent paper from Allen (along with his colleague Marco Pavone) at Stanford's Autonomous Systems Laboratory says.

The database landscape is much more diverse than it once was, thanks in large part to big data, and on Tuesday, one of today's newer contenders unveiled an upcoming release featuring a major boost in security. Version 9 of MarkLogic's namesake NoSQL database will be available at the end of this year, and one of its key new features is the inclusion of Cryptsoft's KMIP (Key Management Interoperability Protocol) technology. MarkLogic has placed its bets on companies' need to integrate data from dispersed enterprise silos -- a task that has often required the use of so-called ETL tools to extract, transform and load data into a traditional relational database. Aiming to offer an alternative approach, MarkLogic's technology combines the flexibility, scalability, and agility of NoSQL with enterprise-hardened features like government-grade security and high availability, it says. Now coming up in the next generation of the software will be a variety of improvements in data integration, manageability and security, the company says, but certainly most notable among them is the addition of Cryptsoft's KMIP.

Microsoft's Azure Machine Learning is one of many changes making its way into Dynamics CRM 2016, the customer engagement software set for release later this year, the company announced Nov. 5. We first learned about the development of Dynamics CRM 2016 in September, when Microsoft promised the upcoming version would be more mobile-friendly and include better transitions across apps like email, Excel, and OneDrive for Business. The push to improve Dynamics CRM continued in late September 2015, when Microsoft acquired technologies and assets from partner Adxstudios. Web portals from Adxstudios are built into Dynamics CRM so sales and customer service can be conducted online; its tools are also designed to improve customer communications. For the upcoming release of Dynamics CRM 2016, Microsoft plans to release related updates in waves as the launch grows closer.

We have been convinced for many years that machine learning, the kind of artificial intelligence that actually works in practice, not in theory, would be a key element of the next platform. In fact, it might be the most important part of the stack. And therefore, those who control how we deploy machine learning will, to a large extent, control the nature of future applications and the systems that run them. Machine learning is the killer app for the hyperscalers, just like modeling and simulation were for supercomputing centers decades ago, and we believe we are only seeing the tip of the machine learning iceberg as Google, Facebook, Baidu, Amazon, Microsoft, and other titans of the Internet, who have enough data to make machine learning not only practical, but necessary, build out their expertise and embody it in the development and production platforms that support their empires. Google's first machine learning platform, called DistBelief, was rolled out in 2011 and used to train deep neural networks using tens of thousands of CPU cores across thousands of servers, By its own admission, DistBelief was difficult to use and tied very tightly to Google's own infrastructure, and so the company created a better, more generic machine learning platform called TensorFlow, which was unveiled and open source last November.