SWAO:Conference Call 20211011

Number 93
Duration 1 hour60 minute
3,600 second
0.0417 day
Date/Time October 11 2021 18:00 GMT
11:00am PDT/2:00pm EDT
7:00pm BST/8:00pm CEST
Convener Andrea Westerinen

IAOA Semantic Web Applied Ontology (SWAO) SIG

THIS MEETING TOOK PLACE ON MONDAY 11 OCTOBER 2021

Meetings are normally on the first Monday of the month at 2pm Eastern Time. This meeting was moved to increase attendance, since various participants were available due to country holidays.

GOING FORWARD ... Due to attendance problems with people's schedules, it was decided to move the meetings to the second Friday of each month, leaving the time at 2pm Eastern.

Agenda

  • Actions from last meeting
    • KenB status update on the workshop communique from the IEEE International Conference on Cognitive Methods in Situation Awareness and Decision Support (CogSIMA)
  • Main points for a analysis of the various definitions of ontologies
    • Decision to write a paper on this topic
  • Housekeeping and next meeting
  • AoB

Proceedings

Actions Review

  • KenB was not in attendance - CogSIMA update still OPEN

New action items discussed below.

Ontology Definitions and Analysis

There was a healthy discussion of these slides. Note that the slides have been edited to include all the feedback received before and during the meeting.

The main discussion points are:

  • The idea of writing a paper originated with a discussion of ontology "styles" by Mike Bennett. That discussion morphed into one about "guidelines". But, before discussing guidelines, one needs to have a definition of what an ontology "is".
    • The latter has been the subject of the last several conference calls.
  • A definite focus is to bring together different communities (as defined by SWAO's charter)
    • SWAO has a focus on "applied" ontologies -> So, this puts a focus on "practical" ontologies
    • "Practical" is defined as "of or concerned with the actual doing or use of something rather than with theory and ideas"
    • This may be "bottom up", "top down" and/or "middle out" depending on how the problem is defined
  • Can we address "problem solving" ontologies -> Ontologies meant to address specific problems and define concepts for specific domains
    • How does the rendering of the ontology play into the problem?
      • Common language vs FOL
      • Can we compare semantic web applications to applied ontology logic applications?
  • "Machine processable" is an important part of a definition of ontology
    • BUT, machine processable does not necessarily require a FOL definition
    • One can follow from the other
  • A DEFINITION:
    • Ontology can be defined as a partial, simplified conceptualization of the world as it is assumed to exist by a community of users
    • A conceptualization created for an explicit purpose and defined in a language
    • Language should be serializable to a machine processable form
  • Important to emphasize that different types of ontologies exist for different purposes, such as vocabulary defn, knowledge sharing, ...
  • Logical formalism helps to disambiguate multiple definitions of terms, different layers of logic
    • And grounds primitive concepts
  • Uses of ontologies aid in interactions with real-world meaning - NLP, data fusion, ...
    • Need to distinguish between concept and data (concept is about defining what something is, vs data being the specific info that some entity has)
    • The data may be a "digital twin" or may only be a surrogate for the concept
  • Reasoning over data - starts from use case with classes and restrictions of the problem space
  • Over the years, much has been discussed about the "semantic ladder"
    • Moving from dictionaries to taxonomies to thesauri to ontology
    • Need to relate the different aspects of the semantic ladder to the themes of SWAO (semantic web and applied onto)
    • Requires different activities across spectrum; Want to use all parts of the spectrum
    • There are more constraints in FOL - really moving to more specific definitions
    • Is this about adding "annotations"?
  • You can annotate concepts and then take in 1 of 2 (or 3?) directions:
    • Probability/possibility are involved - A variety of different things can result from x - depending on the conditions
      • For example, snowfall can melt or can turn to ice or ?? - So, things about e.g snow may not be able to be axiomatized with the expressive power we have available in our ontology language
  • => There are things we can axiomatize, but only in relation to concepts we won't have data for (e.g. social / commitments)
    • Versus things that are axiomatized and for which we have data
    • Similarly, a "taxonomy" of business terms may include axioms but there may be concepts that can't be axiomatized (that need definitions)
      • May have nothing to do with actual data so need human level definitions
  • Can use an ontology to define how things are connected
    • A "coherent descriptional framework"
  • (NEW ACTIONS) RobertR:
    • Liaison with IAOA Education Committee regarding SWAO defining "what is an ontology"
    • Share spreadsheet that is a catalog of definitions of 'ontology', ... / Would only be used with attribution
    • From a quick review of the spreadsheet, questions came up as what a "generic"/general ontology is? Is it top-level or foundational or about abstract concepts?

Next Meeting

The next SWAO SIG meeting will be November 12 2021, 2pm Eastern Time. (Back on standard time, world-wide.) After discussion with the participants, a Friday afternoon call was deemed the best alternative.

The topic of discussion will be a review of an "ontology" writeup (AndreaW to summarize) based on the above discussion.

Attendees

  • Andrea Westerinen
  • Mike Bennett
  • Gary Berg-Cross
  • Robert Rovetto
  • Nancy Wiegand

Previous Meetings

... further results