From OntologPSMW

Revision as of 13:22, 23 October 2019 by KennethBaclawski (Talk | contribs)

Jump to: navigation, search
[ ]
    (1)
Session Introductory
Duration 1 hour60 minute
3,600 second
0.0417 day
Date/Time October 23 2019 16:00 GMT
9:00am PDT/12:00pm EDT
5:00pm BST/6:00pm CEST
Convener Ken Baclawski

Contents

Agenda     (2A)

Conference Call Information     (2B)

Participants     (2C)

Proceedings     (2D)

[12:25] RaviSharma: We are discussing John's slides and the Graphic put together by Ken and Janet to depict the concept of Knowledge Graph.     (2D1)

[12:29] RaviSharma: wiki data is an example of very large knowledge graph - from John     (2D2)

[12:32] David Eddy: If you do a Google search you will be ignoring a huge slice of reality that Google does not slurp.     (2D3)

[12:34] David Eddy: Google does not & cannot adequately search software code since software does not play by the rules of human languages.     (2D4)

[12:41] ToddSchneider: Computers do not assume anything. People make assumptions about a computer.     (2D5)

[12:42] Paul Tyson: D.J. Norman reduces OODA to 2 steps: Evaluate, Execute. That could be taken as the universal prototype for animal activity.     (2D6)

[12:45] RaviSharma: John explained slide 44 Tarski style depicting basics related to KGs as Static and its iteration leads to predict     (2D7)

[12:46] RaviSharma: OODA (Todd described) dynamism is 39.     (2D8)

[12:48] ToddSchneider: Ravi, the OODA loop was originally designed for dynamic 'knowledge'.     (2D9)

[12:50] Mark Underwood @knowlengr: Perhaps an unfair complaint, but one would want to be able to accommodate a graph such as a network map and connect it to an asset inventory represented as an ontology. I'd thought this would be handled as a microtheory within Information technology, but the representations here seem to be upper level abstractions that aren't amenable to connecting assets to network graphics using ontologies     (2D10)

[12:50] Mark Underwood @knowlengr: *graphs not necessarily "graphics"     (2D11)

[12:51] David Eddy: But Wikipedia totally ignores, has no access, to huge parts of "reality." Example being... & I asked repeatedly to folks at MIT SemanticWeb meetings business applications that bring food to the grocery stores.     (2D12)

[12:52] David Eddy: @Ravi... surely you're not looking for a SINGLE definition of "knowledge soup."     (2D13)

[12:53] janet singer: Supporting Johns point re knowledge soup: Wikipedia has a policy that Wikipedia is not considered a reliable source     (2D14)

[12:53] RaviSharma: Discussed Knowledge soup concept.     (2D15)

[12:54] Ken Baclawski: The domain knowledgesoup.com is owned by someone, but knowledgesoup.org is available.     (2D16)

[12:54] MikeBennett: Knowledge = Information + Meaning     (2D17)

[12:55] David Eddy: @MikeB.... good to see "data" is not in that equation.     (2D18)

[12:55] RaviSharma: Real world models are probably not very reliable for many areas, yet they are a source like knowledge soup?     (2D19)

[12:57] RaviSharma: John - Knowledge soup can be all artifacts related somehow and an app or other apps trying to communicate about the subject!     (2D20)

[12:59] MikeBennett: @David indeed. One could argue that data is a kind of information (as are marks on paper, waves in air, DNA encodings etc.). Information at its most general is er Shannon etc. - non randomness. Each of these words return different meanings for different people but these are the distinctions I find useful to work with.     (2D21)

[13:01] RaviSharma: David - KS it is at best a heterogeneous concept like alphabet soup that even contains.... many components and unrelated entities apples oranges horses etc.?     (2D22)

[13:02] David Eddy: "word senses"... and remember that professional lexicographers EXCLUDE stuff like acronyms, abbreviations, etc.     (2D23)

[13:02] David Eddy: And just because a LABEL says one thing... the contents behind the label do not have to agree. Consider the issue of "overloading."     (2D24)

[13:05] Mark Underwood @knowlengr: Have to jump to another meeting. thanks     (2D25)

[13:05] David Eddy: Once again... I urge "unnatural language" to be included here... not just ignored.     (2D26)

[13:05] ToddSchneider: Mark, Have you seen the work done at MITRE on cyber-security and ontologies?     (2D27)

[13:06] RaviSharma: John - said follow up Qs are very imp about KGs. Ravi says as was the case for last summits.     (2D28)

[13:07] David Eddy: @Janet... very hard to elbow my way to this table.     (2D29)

[13:07] RaviSharma: Janet - Precision, rigid, irreducible, KGs to be negotiating between HH Comm vs Computerized, consistent formal W theory.     (2D30)

[13:08] David Eddy: HH = ? Helly Hanson?     (2D31)

[13:08] ToddSchneider: David, would 'formal language' (e.g., FOL) be an 'unnatural language'?     (2D32)

[13:09] David Eddy: @Todd... not a clue     (2D33)

[13:12] Ken Baclawski: MetaNet website: https://metanet.icsi.berkeley.edu/metanet/     (2D34)

[13:29] ToddSchneider: A function 'ontology': https://fno.io/spec/     (2D35)

Resources     (2E)

Next Meetings     (2F)