From OntologPSMW

Jump to: navigation, search
[ ]
    (1)
Session The Ontological Landscape
Duration 1 hour
Date/Time 21 Apr 2021 16:00 GMT
9:00am PDT/12:00pm EDT
5:00pm BST/6:00pm CEST
Convener Todd Schneider
Track A

Contents

Ontology Summit 2021 The Ontological Landscape     (2)

Ontologies are a rich and versatile construct. They can be extracted, learned, modularized, interrelated, transformed, analyzed, and harmonized as well as developed in a formal process. This summit will explore the many kinds of ontologies and how they can be manipulated. The goal is to acquaint both current and potential users of ontologies with the possibilities for how ontologies could be used for solving problems.     (2A)

Agenda     (2B)

  • Commercial Ontology Development Michael Uschold (Semantic Arts) Slides Video Recording YouTube Video     (2B1)
  • Abstract: We report on our experiences in developing enterprise ontologies for deployment in production systems in a wide variety of industries. We describe our "Think Big - Start Small" approach which avoids creating semantic silos and ensures smooth adoption across the enterprise. A critical enabler is our use of gist, an upper enterprise ontology. We describe how our overall approach has evolved over time. Increasingly, we are being nudged by our clients to reuse existing ontologies and vocabularies. There are many to choose from. It seems like a great idea, but there be dragons.     (2B2)

Conference Call Information     (2C)

Attendees     (2D)

Discussion     (2E)

[12:09] RaviSharma: Michael - how do you handle duplicated triples among different silos and reconcile if their meanings are different slightly or vastly?     (2E1)

[12:11] RaviSharma: thanks for answering that by generating a schema (new).     (2E2)

[12:19] RaviSharma: How many iterations with SMEs from different areas of enterprise are required to reconcile Think Big items?     (2E4)

[12:21] RaviSharma: internal reuse - When you borrow chunks of OWL you get URLs but how do you know content is the desired pattern, does that form your authoritative source of Info, I would think chunks mean differently in different enterprises?     (2E5)

[12:23] RaviSharma: enterprise ontology is by Domain such as electric, manufacturing etc.?     (2E6)

[12:24] RaviSharma: does temporal relationship help you with provenance and thereby residual contract and warranty aspects, thru ontologies?     (2E7)

[12:25] RaviSharma: Michael - very interesting phenomenological approach and flexibility!     (2E8)

[12:30] Gary Berg-Cross: To discuss the ontology landscape we discuss the landscape of the various ontology/knowledge engineering practices that produce them. Some are less formal and mature than others.     (2E9)

[12:31] RaviSharma: How much is "gist" and what else remains to create enterprise ontology?     (2E10)

[12:37] RaviSharma: Michael - if FIBO had strong TLO, would that have helped in your middle block taxonomy layer's financial items?     (2E11)

[12:37] RaviSharma: on slide 23.     (2E12)

[12:42] RaviSharma: How do you associate Shapes to Data to enable use of SHACL     (2E13)

[12:43] Mark Underwood: Are you finding any help in reuse from external non-ontology sources, especially enterprise CMDB's (eg ServiceNow CMDB) or IT standards such as DMTF Redfish?     (2E14)

[12:44] RaviSharma: Since implementation of software does not follow say CEO era, how do you deal with both provenances?     (2E15)

[12:48] RaviSharma: From what you describe, a lot of manual intervention is implied in this effort, how data stewardship coexist with data curators e.g.?     (2E16)

[12:48] BobbinTeegarden: Do you address the 'process' side, or is that an exercise left to the client?     (2E17)

[12:52] TerryLongstreth: Slide 33; is using the intermediate triples better than improving physical structure of the RDB (adding or changing indexes, hashes, hierarchies)?     (2E18)

[12:52] Gary Berg-Cross: We can perhaps spend a little time in our synthesis considering the landscape of pseudo-ontologies. Some are used for technical convenience as part of some process such as query efficiency. There are probably several examples as part of building, using, and maintaining KGs.     (2E19)

[12:54] janet singer: As I recall from the Knowledge Graph seminar at Stanford last year, you said that you prefer speaking of generic conceptual schemas rather than ontologies. Are the continued attempts to popularize the ontology label a matter of paving cow paths that happened to be laid decades ago? Would development of culture and technology for eliminating semantic silos be better served with some other language?     (2E20)

[12:56] RaviSharma: Michael said there is company formed by Juan Sequeda that calls it depleted ontology - in answer to Gary's Q.     (2E21)

[12:56] Mark Underwood: @Ravi "Putative" not depleted lol     (2E22)

[12:57] Mark Underwood: yours is more colorful     (2E23)

[12:58] Cas Miles: Thank You!     (2E24)

[12:58] TerryLongstreth: Uschold's answer was non-responsive. I'll send him an email.     (2E25)

[13:09] Gary Berg-Cross: The Century Dictionary, http://www.global-language.com/CENTURY/, a massive reference work (first edition 1889, Supplement 1909, final edition 1914) contains many thousands of definitions by Peirce. The words he defined are listed alphabetically on the website of the Université du Québec à Montréal, where a branch of the Peirce Edition Project is preparing a volume of the Writings (W7) entirely devoted to those definitions. For an article on Peirce's Century Dictionary work, see the Peirce Project Newsletter 3:1 (1999). The entire Century Dictionary is now online in a searchable format. Cited by title of entry in quotation marks.     (2E26)

Resources     (2F)

Previous Meetings     (2G)


Next Meetings     (2H)