Gunning the Microsoft Semantic Engine

New Bing Maps Beta with embedded data layers from Twitter and other social feeds, click to enlarge screenshot

There’s a lot of information on the Internet already. Every day, more is added – a lot more. And while there are a concomitant number of new analytic or sense-making tools on the web, they butt up against the fact that the data – the all-important data – is held in multiple places, formats, and platforms.

How are we going to deal with all this? One approach is almost mechanical: ensuring that datasets can be accessed commonly, as in our new Microsoft Dallas platform associated with the Windows Azure cloud platform.  In the government realm, the anticipated reliance on “government-as-a-platform” (a meme popularized by Tim O’Reilly) holds promise in allowing somewhat aggregated datasets, openly accessible.

Another approach is at the far end of utility, on the post-analytic end with better search and decision engines. That’s why we’ve just upgraded Bing with some phenomenal features (see “Bing Maps Spawn the ‘Decade of the Interface?’ IDC thinks Bing Maps is on the start of big advances in Web user interfaces“).  In a neat demonstration of the value of competition, by the way, Google just followed suit a week later with some strong new features as well.

But an additional, and powerful, approach in the middle of the data value-chain will be to use semantic techniques on all of the data itself. A couple of weeks ago we unveiled the Microsoft Semantic Engine, which brings search, structured query, and analytics together in a simple, unified Search-Discover-Organize usage model.

Having all forms of digital information on a single platform allows people to blend unstructured and structured content and to drive insight and decision making

 Microsoft Semantic Engine provides a combination of technologies to form a contextual understanding of all digital content


MSE has been designed to allow unified access to structured and unstructured enterprise data through easy-to-use analytical tools. Microsoft might build some of those tools, and we’ll certainly integrate MSE capabilities into our own “information-worker” software, just as it is going to be integrated into Windows 7 desktop search and other apps. But the platform is meant to be easily available for use by the world of developers and partners, so that their independent innovations can take advantage of the semantic engine.

MSE Architecture, showing array of pluggable services including inference engines, ontology and taxonomy managers, image and text processors etc. Click to expand screenshot.

Therefore we held the unveiling at our Professional Developers Conference in Los Angeles. While MSE is not available for use today, it’s right around the corner, so we are busily providing information to the developers and partners whom we hope will be using the engine.

You can get the same experience that PDC attendees had, by watching the video of the presentation and demo here, or downloading the slides which accompanied the roll-out and live demo. The video itself is about 45 minutes long and features talk and demoing by Naveen Garg and Duncan Davenport of the MSE team.

I first met Naveen two years ago, not long after joining Microsoft, when I began hunting for smart people working on semantic projects like one I had just left in the intelligence community – our “Alien” project (All-Source Intelligence Environment).  I’ll point out that at 10:30 into the video, Naveen speaks for a moment or so about the interest within government circles in our semantic approach; he is referring to discussions we have had with visitors I’ve brought to Redmond from Washington DC for a look at the engine in its early lab days. They enjoyed meeting Naveen and his team, and I have to say they’ve cut us a bit of slack as the MSE team hunkered down and focused quietly on their work.

MSE is not yet a “product” for sale by Microsoft, and may not become one per se, at least by itself. We’re still going through the process of figuring out exactly how it will be packaged. But the rationale for surfacing it now is that we want to encourage developers who are already working in knowledge-rich corporate environments or other information-centric arenas, to be prepared for the ability to take control over structured and unstructured content, in a relational way that will provide contextualized, personalized insight into massive data for users – whether they’re web surfers or enterprise users.

Here are some of the scenarios we believe are relevant for using MSE:

  • Search and Collaboration | Personalized search, discovery and organization
  • Legal | Precedent and subject based search over large scale textual corpuses
  • Life Sciences | Systems biology with large volume data correlation and search
  • Government Services | Intelligence, real-time analytics, visualization, clustering
  • Social Networking | Social graph relevance mining, ranking criteria
The video makes clear the PDC talk was targeted to developers, with the main discussion about the scalable architecture which can allow mulitple text processors, image processors, video and object processors – we actually provide a full stack of managed RESTful APIs, web-services, and even T-SQL syntax over all the entity properties that we extract.

MSE Demo, with screenshots of Search, Discover and Organize features, exposed via UX gallery using fully-documented managed APIs. Click to expand screenshot.

Just a couple of highlights in the video to watch for: At 36:15 you’ll see an integrated MSE semantic search within Windows 7 Desktop Search, which shows the underlying SQL Server search going on, with all semantic properties exposed and analyzed automatically for relevance and semantic relationships within the Desktop Search UI.

And at 39:30, there’s a quick 3-minute technical discussion, describing how we’ve integrated the Semantic Engine Database, the System Integration Fabric, the API Layer, and the entity processors and classifiers against different data types.

If you want a bit more technical take on the presentation, check out this blog from a SQL Server business-intelligence (BI) developer, who is excited about the possibilities he sees from using MSE.

I’m excited too – I’m a big believer in Our Glorious Semantic Future and I believe that Microsoft now has a firm footing from which to build great capabilities. If you want to stay current on our progress and plans with MSE, feel free to drop me a line.

Share this post on Twitter

Email this post to a friend

AddThis Social Bookmark Button

3 Responses

  1. […] here: Gunning the Microsoft Semantic Engine « Shepherd's Pi By admin | category: engine search, sem | tags: microsoft, microsoft-semantic, sem, […]


  2. Glad to see that MS is taking the semantic web is heading mainstream. Across the world people will be debating entity resolution – Nice blog


  3. […] semantic techniques and power. It was just six months ago – but seems like years – that I wrote about our work with our Microsoft Semantic Engine, its REST API layer, the pluggable services including inference engines, ontology and taxonomy […]


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: