T3DD17: What is nreach?

During this year TYPO3 Developer Days (T3DD17) in Malmö, Schweden Johannes and I had the chance to present our work on nreach. This posting recaps parts of the presentation and our next possible steps for interested developers and agencies.
You can directly go to the slides and videos or read this summarization.

Which problems are solved by nreach?

Finding the right information is still a difficult task:

  • It's time-consuming and not efficient
  • It requires a great amount of knowledge on
    • the correct key words
    • the context
    • the tool
  • Extra: If the system knows what it is handling, it can do a lot for you
How are these problems related to TYPO3?

Website visitors want to find information. Website creators also need to find information to be able to provide it on their website. So TYPO3 is directly related to the problems above.

What is the future vision?

nreach offers products that enable you to add value to your content management system. This happens through the use of state of the art technology cognitive technologies:

  • Computer vision
  • Text analysis and enrichment
  • Metadata modeling
  • Knowledge discovery
  • Visualisation

Nreach provides simple integration of these services into your CMS:

Which concrete products does nreach have in mind?
Computer Vision

Our computer vision integration uses Microsoft Azure cognitive services to add the following enhancements to your images in TYPO3:

  • Description of the image
  • Keywords
  • Face recognition
  • Gender and Age of person
  • Emotional analysis
  • Colors in use
  • Probability of detected concepts
  • Meta information such as photographer, copyright and camera info

See some example videos:

Text analysis and enrichment

Our text analysis and enrichment component based of CKEditor integrates the Microsoft Azure Entity Linking Intelligence Service-API with TYPO3:

  • Named entity and recognition
  • Context awareness and disambiguation
  • Linking of named entities to external sources
  • Microformat support (RDFa, Schema.org, Web Annotations)
  • Search engine optimization
  • Cognitive enhancement for Solr Facets

It also allows one-click translation between different languages while preserving the HTML structure.

Metadata modelling

Our metadata modeling module allows you to create a sensible set of information descriptors. Using schema.org entities in TYPO3 to
Make your content understandable for search engines, browsers, and apps.
For a deeper introduction see our earlier blog posts 1 and 2.
This will later allow you to:

  • Have a reliable metadata collection as knowledge source
  • Compare your content with other content (content gravity)
  • Recommendation of related and relevant information
  • Search for unknown terms using similar high-level entities
Knowledge discovery

Our knowledge discovery options allow accessing quantitative and qualitative aspects of your information collection by applying AI and ML.

  • Who has created content
  • which impact on information performance
Visualisation

An image replaces a thousand words. Our visualization component will enable you to recognize the deep value of your information:

  • Triplets
  • Relation graphs
  • Tags and ontologies
Nreach Architecture

nreach is not a “pure” TYPO3 extension, that would hardly be possible.
We are using a service architecture and rely on remote calls.

Try it out

Detailed installation instructions Basically two steps:

  • composer require nreach/nreach-t3-vision
  • Sign up
  • Configure your key inside the extensions configuration within the extension manager
Two more things

Chat integration into TYPO3:

nreach integration with solr:

nreach website nreach developer portal nreach github

von Olivier Dobberkau

Olivier Dobberkau is founder and CEO of dkd Internet Service GmbH, and President of the TYPO3 Association. He has over 17 years of experience in new media and the internet.