Skip to main content

Cloud

Sitecore Symposium 2017: Cognitive Services, Deeper Intelligence

Microsoft Cognitive Services provides a wide range of cool tools. They are all available via API. Sitecore continues to evolve. It’s no longer just a content management system. It’s evolving into a content learning system.
The Sitecore Symposium 2017 presenter took an image and uploaded it and had it analyzed. It understood that it was a female about age 24. It unlocks your Media Library. Now you can do a faceted image search. You can search, filter, explore, match content, etc. Content editors should rejoice.
There are a number of use cases:

  • Manage User Generated Content
    • Automatically filter out violence, nudity, etc.
    • Be smart enough to initiate a human review
  • Custom Vision Service
    • Upload labeled images
    • Identify logos
    • Help the tool to continually improve on your content
  • Language Parsing
    • Parsing
    • Language detection
    • Translation
    • Transcription
      • Audio to text
      • text to audio
      • transcribe video
  • Bot Framework
    • Enhanced search
    • Multilingual support
    • Accessibility implications, better support those with special needs
    • Sitecore now has “Ole” This is like Cortana and Alexa.
  • Recommendations Engine
    • Recommended Content
    • Get better customer engagement
    • Do commerce integration with it
  • Labs in process
    • Location, events,
    • Time, distance
    • Routing logistics
    • Gesture Controls

Demo Time

The presenter did this from Sitecore:

  • Start in the Media Library and upload an image.
    • On upload, it does analysis
    • It capture faces in visual analysis. Includes confidence indicators
      • Also will tell you if it’s “adult” or racy content
      • Tells you the colors. Imaging now searching for pictures in a color scheme
      • Can learn well known and not so well known people and identify them
    • Emotional analysis: Is it angry, happy, sad, etc.?
    • Facial analysis: show facial hair, etc. (lots of metadata here)
    • Textual analysis: Read text out of an image. It knows where and what the text is.
      • The image text is now searchable.
  • Jump into experience editor, open a rich text field and add an image
    • Pop into image search.
    • Do a lot of searches.  They were really well tagged.
    • He then searched by facets:
      • Size, gender, glasses, age, emotion, etc.
      • The facets are VERY interesting.
  • Ole – the new chat tool
    • He asked Ole what version he is. (Sitecore v 8.2)
    • You can still publish content.
    • This is different from command line. It’s a conversation. You can ask the same question multiple ways with slightly different syntax and it will get it right.
    • He showed a publish use case.  He gave the path, channel, database, etc. and voila, it’s published.
    • Also asked about how many locked items were there and then unlocked five items.
    • Think about any user just telling the system to do something. That’s a huge change.

What Is Involved?

This is a true framework and not a hack job. Mark Stiles wants to let us know it’s available now on Git Hub:

  • GitHub.com/markstiles/SitecoreCognitiveServices
  • markstiles.net/blog
  • @markstiles

It uses the Helix architecture and he has a repository for each API. In the project layer, he provides code samples to help you get started.

Road Map

  1. More chat capabilities. It’s his focus right now. Make it more useful
  2. Marketplace modules
  3. TBD. Would love some input
  4. Eventually makes profit from it 😉

The Logo

And yes, Mark even has a logo for this. I’ll let you get the reference to an old Scifi TV show….

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.

Michael Porter

Mike Porter leads the Strategic Advisors team for Perficient. He has more than 21 years of experience helping organizations with technology and digital transformation, specifically around solving business problems related to CRM and data.

More from this Author

Follow Us