Actions

Ontolog Forum

  • Ideas and preparation:
    • Intent today is to review project status and current challenges
      • Review Goals, Resources, Expectations and Expectations for next 6 weeks (July 23rd Face to Face)
      • Consider User Profiles, Perhaps a Use Case
    • The five phase RoadMap discussed at the last Taxo-Thesaurus (T-T) work session
    • Objectives of this phase: Domain Maps and Respective Perspectives
    • Common or Shared Meanings, Layers, and Lacunae (aka Gaps and Holes in the Fabric of Meaning)
    • Suggestive Questions about: Intermediate and Intended T-T Product, T-T Process, T-T Tools, T-T Risks
    • Etc.

Conference Call Details

  • Date: Thursday, June 15, 2006
  • Start Time: 10:30 AM PDT / 1:30 PM EDT / 17:30 UTC (see world clock for other time zones)
    • Duration: 1.5~2.0 hours
  • Dial-in Number: +1-641-696-6600 (Iowa, USA)
    • Participant Access Code: "686564#"
  • Shared-screen support (VNC session) will be started 5 minutes before the call at: http://vnc2.cim3.net:5800/
    • view-only password: "ontolog"
    • if you plan to be logging into this shared-screen option (which the speaker may be navigating), and you are not familiar with the process, please try to call in 5 minutes before the start of the session so that we can work out the connection logistics. Help on this will generally not be available once the presentation starts.
    • people behind corporate firewalls may have difficulty accessing this. If that is the case, please download the slides below and runing them locally. The speaker will prompt you to advance the slides during the talk.
  • RSVP to peter.yim@cim3.com appreciated, to allow us to prepare enough conferencing resources.
  • Please note that this session will be recorded, and the audio archive is expected to be made available as open content to our community membership and the public at-large under our prevailing open IPR policy.

Attendees

  • Also Expected (people who mght have joined after the roll call):
    • Patrick Durusau
    • James Werner
    • one more colleague of Michael Uschold (--please provide full name and affiliation)
    • Raj (rajupvk) (--please provide full name and affiliation)
    • Paul Koch (may join late)
    • ...(to register for participation, please add your name & affiliation here or e-mail <peter.yim@cim3.com> so that we can reserve enough resources to support the session.)...

Background

Ontolog Forum is rolling out a series of events (talks and discussion sessions) between April and July 2006 that revolve around the topic: "Ontologizing the Ontolog Body of Knowledge" during which this community will explore the "what's" and "how's" to the development of a semantically interoperable application, using the improved access to the community knowledge of Ontolog as a case in point.

The OntologTaxoThesaurus project is one of the activities spurred by the "Ontologizing the Ontolog Body of Knowledge" exercise.

At this session Professor Bob Smith (Cal State U & Tall Tree Labs), Dr. Denise Bedford (World Bank) and the OntologTaxoThesaurus team will be running a combined work session and project review session on Thursday June 15, 2005 to allow knowledge engineers and domain experts to confer and help categorize 'concepts' extracted from Ontolog's knowledge content, as part of the effort in the overall OntologTaxoThesaurus project.

Agenda & Proceedings

Topic: OntologTaxoThesaurus Project Work Session - Categorizing Concepts

  • Abstract (by Bob):
    • This Ontolog Taxo-Thesaurus Project is documenting a process for "Building a Top Down Ontology from the Bottom Up". Our primary goal is an easier and more comprehensive access to Ontolog's Knowledge Base. Our focus is on the extensive heterogenous body of knowledge contained within this Wiki. The Project Team of volunteers is pursuing the RoadMap describe by Dr. Bedford. This RoadMap includes 6 sequential Steps using tools and judgments to produce documents such as Concept Lists, Concept Clusters, Category Profiles, etc.
  • Session Format & Agenda: this is be a virtual session conducted over an augmented conference call
    • Introductions
    • Ontolog's Taxo-Thesaurus Project Overview
      • See Figure 1, below - http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2006_06_15#nidNRR
      • Concensus Objective is to improve Search Process with taxonomic and thesaurus based terms and concepts
      • Methods involve a process we now call Bottom Up development of a Top Down Ontology
      • Steps or stages involve inventory of bounded content (about 65+ items with over 10K terms)
      • Concept extraction, concept clustering, and domain expert discussions
      • Category profiles and improved specification of content
    • Status Report
      • Work completed to date
      • Work to be accomplished today
      • Future tasks
http://ontolog.cim3.net/file/work/OntologizingOntolog/TaxoThesaurus/taxo-thesaurus_steps--DeniseBedford-BobSmith_20060612a.png
  • Today's Discussion:
    • Denise Bedford: have a list of about 11,000 terms which could be build into our controlled vocabulary
      • one goal: someone new to the technology can come to the Ontolog site and learn about the concepts (even as a novice)
    • Pat Heinig: how does the process map come by
      • Bob Smith: using a tool call 'Mindmap', which is based on XSLT
        • the intent is to keep things on one page
    • Bob Smith: another objective is to get an 'improved search' over the Ontolog content
    • Denise Bedford: the process (see: slides)
      • (using COAST) we inventoried about 65,000 objects from the Ontolog content
      • capturing only noun phrases, we came up with a list of about 15,000 terms
      • manually look for clusters of concepts
      • that could be aggregated into categories
      • which we will then present to domain experts - asking them: 'what's missing?' and 'what's not pertinent?'
        • which, in our case now, are domain experts in 'ontologist' - but important to go back onto the questions of 'who the users will be?', and 'how are they going to use it?'
      • then, prune the list
    • Dagobert Soergel: suggest we use 'facets' and put up a search engine that supports it
    • Bob Smith: we want conceptual alignment up front
      • Pat Heinig: suggesting the use of some consensus or harmonization mechanism, and apply it over the wiki
        • more classic processes like 'Nominal Group Technique' or "Delphi' may be adapted into our wiki-based workspace (help improve the signal-to-noise ratio)
      • Kathleen: where do we start, e.g. in handling disagreements
        • Denise Bedford: we might first try to just decide whether an item should be 'included' or 'excluded'
      • EMichaelMaximilien: maybe we can use 'rating' ... like how SLASHDOT does it, or the way DIGG applies to News items
      • Charles Turnitsa: from experience, a lot of disagreements can rise because its complexity ... breaking it down to sub-concepts sometimes will bring universal agreement
    • Denise: Other goals - (a) (ref. Peter) building an ontology of the Ontology domain; and (b) (ref. Bob) tracking the process and learning from that.
    • EMichaelMaximilien: how do we expect to continue building, maintaining, a dynamically changing body of content
      • Denise: we give the 'cloud' to the community ... and will need to continuously organize (and maintain) it
    • One other question we can ask everyone is "What other sites should we be looking at for domain concepts?"
    • Bob: how many within this group are comfortable editing this wiki?
    • Bob: we'll route the list of concepts to those who are interested
  • Additional Resources:
    • (please provide additional material & input here)

Questions, Answers, Discourse & Notes

  • (please insert here)
  • For those who have further questions and discussion on this topic, please post them to the ontolog forum so that we can all benefit from the discourse.

Session ended 2006.06.15 12:06 pm PDT