Ben  Franklin Ben Franklin | 27 Apr 2020

Since deciding to make the move to Azure a few months ago, Quba has been implementing some of the features that Azure offers. The extent of Microsoft’s Infrastructure-as-a-Service, as it calls itself, is much more than just cloud hosting. Azure has in fact some very clever tricks up its sleeve, and they can become an important set of tools in your comms strategy kit if you know how to make best use of them.

When we say clever, we mean proper brainy. In Part 1 of this series, we’re looking at Azure’s cognitive services. Microsoft has really invested in its artificial intelligence, and Azure is using it to put quality analysis tools into users’ hands. Azure assembles years of Microsoft research in text, image, and language analysis and recognition, and it has incorporated this knowledge base into its search function.

Image analysis can identify all the usual basics, such as colours and whether there are people present. But Azure can also identify many well-known faces and brands by name, and it can synthesise all of these data points to give a general description of what’s going on in the image.

Azure’s text analysis AI has the ability to pull out key phrases, much the way that keyword density scorers do. But it also identifies other key entities such as geographical locations, people’s names and titles, times of day or year, and official organisations like the NHS. Azure also has optical character recognition (OCR) capabilities, a kind of a hybrid of text and image analysis. As libraries of content grow exponentially, ever more advanced analysis of this kind can help keep it readily searchable without the need for constant retagging.

Perhaps most interestingly Azure’s cognitive services can give a text an overall mood score. This ranks the text on a scale according to whether its tone and subject is mostly lighthearted and happy or upsetting. This has some obvious benefits for filtering material by appropriateness for audiences. It also has a lot of potential for incredibly refined automated marketing or curating content for personalised user journeys based on certain behaviours. 

It’s already stacked with loads of information, but the real benefit of Azure’s AI is the access to it. It’s there and ready to learn. You can feed it industry or organisation specific data and set the search parameters and rules, and off it goes. The result is Azure Cognitive Search.

When deployed on a website, it means that when searching your database, it doesn’t just trawl through text and tags to come up with search results. It analyses all of the content on your site, from text to video to photos and PDFs, and more, and using its continually evolving knowledge to generate better results every time.

And this is just the start. In Part 2, we’ll talk about how Azure uses these same AI capabilities to provide an excellent bot service. And we’ll take a look at what their built-in security features look like.

If you’re interested in how Azure cloud hosting and Infrastructure-as-a-Service could benefit your organisation, please get in touch.