Business Semantic Interoperability – Part 4 (Final)

In my last post, Business Semantic Interoperability – Part 3, I mentioned before that the preliminary work was almost completed to get the LinguMatic project off the ground. I now have enough content added to the two systems to move forward announcing publicly the availability of the demo systems. It will help me in finding out if there is an interest in making this an Open-Source project.

Here is a summary of what I did to create the online demo systems:

  • There are two online systems using the iQvoc open-source vocabulary management tool that combines easy-to-use human interfaces with Semantic Web interoperability:
    • The main system is for the Business Semantics Thesaurus, which is called LinguMatic. It is the center of the project to provide a solution that allows users to search for their domain specific (everyday) name, resulting in being presented with the definition and any existing related alternate name(s) within their own domain and links (matches) to relevant published standard eBusiness directories/libraries. The URL for the system is http://www.lingumatic.net. The system supports currently two main languages, English and German.
    • The second system, called eBiz – Semantic Directories, is used as a demo system that contains a small subset of the UN/EDIFACT directory, TDED directory and the Core Component Library (CCL). The URL for this system is http://www.ebiz-directories.net. The system’s content is currently only in English, which should not be an issue since the UNECE released directories are only in English as well.
  • The reason for the second system is that the UNECE (UN/CEFACT) doesn’t provide a web-based versions for their directories/libraries containg unquie links to each concept. However, to show how the real-world semantic concepts in LinguMatic map to the standards-world created semantics it was easier to create the second system than to wait for the UNECE/CEFACT to make it happen. Maybe this needs to be part of the project as well.
  • To help demonstrate the functionality of the system as related to the real-world semantics and mapping to the standards-world concepts, I chose the Postal Address as example. You will notice that the narrower concepts of address are also contained, that is to demonstrate how the narrower and broader terms are referenced within SKOS. The Top concept for the LinguMatic system is the Core Thesaurus term, and for the eBiz – Directories there are three Top terms, namely the Trade Data Elements (UNTDED/ISO7372), UN/CEFACT Core Component Library and UN/EDIFACT Directory.
  • The LinguMatic system contains all the postal address terms as defined by UPU in S42-1. The complete hierarchical structure is currently:
    • Core Thesaurus
      • Address
        • Electoral Address
        • Email Address
        • Postal Address
          • Postal Address Segments
          • Postal Address Constructs
          • Postal Address Elements
        • Street Address
  • The eBiz – Semantic Directories’ structure is:
    • Trade Data Elements (UNTDED/ISO7372)
    • UN/CEFACT Core Component Library
    • UN/EDIFACT Directory
      • Address Segment Concepts
      • Address Composite Element Concepts
      • Address Element Concepts
  • The LinguMatic system only has narrower details for the Postal Address, whereas the eBiz – Semantic Directories system has narrower terms for all broader concept in the hierarchical structure.
  • Both system also contain collections that are either a set of concepts or concepts and collections.
  • For now only the eBiz – Semantic Directories system has links (matches) to the LinguMatic concepts and/or collections. The reason being that for now to demonstrate the idea of linkage this is good enough, adding the links back from the LinguMatic concepts to the various eBiz Semantic- Directories’ concepts is a lot of work and assumes that all entries are correct. I like to leave this as a task to the members of the project. There is also the somewhat open issue, what exactly are the exact definitions for the various types of matches (links)? 

There is enough content for now to allow users to play around with and get an understanding of the goal for the LinguMatic system.

That’s it for the final update, I hope that some of you will find time to visit both systems to find out for yourself what it is all about. If you like to participate, drop me a line using the Contact Me page.

This entry was posted in Semantics and tagged , , .