From OntologPSMW

Jump to: navigation, search
[ ]
    (1)
Session Definitions
Duration 1.5 hour
Date/Time 28 Apr 2021 16:00 GMT
9:00am PDT/12:00pm EDT
5:00pm BST/6:00pm CEST
Convener Gary Berg-Cross
Track B

Contents

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)

Conference Call Information     (2C)

Attendees     (2D)

Discussion     (2E)

[12:05] ToddSchneider: Why do we have semantic silos? People.     (2E1)

[12:10] ToddSchneider: What does 'harmonization' consist of or entail in the context of information? What is the scope or context(s) of 'harmonization'?     (2E2)

[12:17] RaviSharma: Gary - there is physics science and otherwise semantic relationships beyond glossaries- related terms in a domain or model?     (2E3)

[12:19] RaviSharma: Gary - Earth sciences domains will describe this similar to tsunami, hurricanes, etc?     (2E4)

[12:20] MikeBennett: @Ravi this relates to the question of whether the meaning inheres in logical intensions (like physics equations and like FOL statement) or inheres in words (Glossary). People sometimes seem to conflate these.     (2E5)

[12:23] RaviSharma: Gary and Mike - this would be interesting if these knowledge related domain terms are woven in an ontology?     (2E6)

[12:25] RaviSharma: Gary - you finally showed use cases of fires and models but also showed possible reuse of ODP!     (2E7)

[12:27] RaviSharma: Why is NASA ESDIS or Tera Chem Aura satellite data and image products are missing glad NOAA is not?     (2E8)

[12:28] DaveWhitten: Gary just talked about Stakeholders. It is important to define areas of knowledge and control by Stakeholders. These aren't quite security issues. But how do we tie those responsibilities, to data fields, predicates, database, data types, hierarchy?     (2E9)

[12:29] TerryLongstreth: How about extended meanings like simile, metaphor: Covid spreading like wildfire, My old flame?     (2E10)

[12:30] janet singer: It what would the competency perspective say about definition of harmonization?     (2E11)

[12:30] DaveWhitten: How do you mark the ontology to say which sources/persons are allowed to be a "definitive source" for this knowledge?     (2E12)

[12:30] RaviSharma: harmonization discussed between Todd and Gary saying parameters in fires are understood and their categories. Aligned in some way!     (2E13)

[12:32] DaveWhitten: The scope/context for information may include qualifications of a person/SME (subject Matter Expert) ?     (2E14)

[12:33] DaveWhitten: I think of an ontology as including the valid ways to draw conclusions. How does this interact with harmonizations?     (2E15)

[12:34] BobbinTeegarden: Gary you seemed to imply a two-level context: a scope within a domain. Nice. Is this heading towards something like a 'fractal' or holonic context?     (2E16)

[12:35] Robert Rovetto: Re: The topics of alignment, matching, mappings, harmonization, and cognates, have some definitions in the literature. There's some nice books/anthologies on it, where they are described in the ontology engineering lifecycle. What is important is explicitly specified the degree of harmonization/matching/... (because it may only be partial), and the type. If one ontology imports more abstract ontologies, then that adds more baggage to be examined and more semantics that are not inherent to the words/terms being defined. Harmonization is not truly achieved if all degrees of abstraction are not examined.     (2E17)

[12:39] Robert Rovetto: no question, just a point     (2E18)

[12:40] BobbinTeegarden: That was great, Gary, combinations of patterns are almost 'parallel' context while a scope within a domain is 'vertical'. How do you automatically morph into the new contexts, is that something you're addressing?     (2E19)

[12:42] RaviSharma: Practical Lexicography would be like Colan classification of Ranganathan?     (2E20)

[12:42] RaviSharma: books categorization?     (2E21)

[12:43] Gary Berg-Cross: @Bobbin I would say there are a few ways we flesh out related contexts. Domain experts will be one source saying things like 'Well there is a climate context to a wildfire." Glossaries developed by the experts also tell us about the contexts in which they discuss things.     (2E22)

[12:45] RaviSharma: Ellipsis would imply predefined Context?     (2E23)

[12:47] RaviSharma: If the second sentence was not there then it could also mean that when i shot elephant i was wearing my Pyjamas?     (2E24)

[12:48] MikeBennett: Another one that often comes up when trying to get info from SMEs is synecdoche. People using a word for a part the whole and vice versa (again because context is assumed).     (2E25)

[12:50] janet singer: Peirce's contributions to the Century Dictionary start on p. 44 at https://www.depts.ttu.edu/pragmaticism/collections/works/bibliography.pdf (NOTE it's VERY slow to load)     (2E26)

[12:51] janet singer: Century Dictionary is at http://www.global-language.com/century/     (2E27)

[12:51] RaviSharma: Shimmer is like scintillation of stars due to atmosphere?     (2E28)

[12:52] Bruce Schuman: "My old wildfire" could be stretching the metaphor - the relationship was a bit more extreme or risky (?)     (2E29)

[12:52] MikeBennett: Some uses are part of creative writing, e.g. P G Wodehouse has Jeeves talk about editing Wooster's face.     (2E30)

[12:55] BobbinTeegarden: Does the pattern matching always involve metaphor or simile?     (2E31)

[12:56] BobbinTeegarden: Similarity is a surprisingly hard problem.     (2E32)

[12:56] TerryLongstreth: @Ken - p14 - "mapping bits" presupposes digital representation     (2E33)

[12:57] RaviSharma: Ken- the talk by Gary showed us that understanding of phenomena for a purpose can use ODP and that purpose can reuse ODPs and even ontologies. In other words fire mitigation implies consistent or Harmonized Ontologies at different levels?     (2E34)

[12:58] janet singer: Ravi: an interesting elaboration. Semantic scintillation results from perception through contextual atmosphere?     (2E35)

[12:58] RaviSharma: Ken - How does that tie with lexicography?     (2E36)

[12:59] Bruce Schuman: Yes, on similarity being challenging. I'd like to see a dedicated study on that question -- I am looking at it right now -- partially informed by Peirce.     (2E37)

[13:01] RaviSharma: Ken - excellent last point?     (2E38)

[13:04] RaviSharma: Janet Thanks!     (2E39)

[13:04] BobbinTeegarden: Having written operating systems, mapping bits works there too... ;0)     (2E40)

[13:10] RaviSharma: Selja said Lexicography and ontology span the two ends.my wording!     (2E41)

[13:11] Gary Berg-Cross: One comment on Exploitation Rules They can be understood in terms of cognitive semantics such as discussed by Lakoff.     (2E42)

[13:12] TerryLongstreth: Lex; fire - V. = (rapid oxygenation, discharge from work, discharge a gun, ...     (2E43)

[13:13] janet singer: As David Eddy would point out we do need to deal with legacy meanings as well, so the evolution of terms over time is important     (2E44)

[13:14] RaviSharma: Selja further described use of term? Ken responded with versioning of ontologies.     (2E45)

[13:14] Gary Berg-Cross: Cognitive semantics uses the idea of an image schemas (type of mental model) that form from experience. These can then be elaborated through embodied metaphor process to structure our problem solving as what we label "abstract" thought.     (2E46)

[13:14] TerryLongstreth: c/oxygenation/oxidation/     (2E47)

[13:15] RaviSharma: MY Q to Ken is whether the entities extracted using Lexicography would differ from those using usual NLP?     (2E48)

[13:16] TerryLongstreth: Ken&Selja - Critiquing ontologies is easy; just change the context     (2E49)

[13:17] RaviSharma: Terry -     (2E50)

[13:21] janet singer: From https://ncatlab.org/nlab/show/red+herring+principle The mathematical red herring principle is the principle that in mathematics, a red herring need not, in general, be either red or a herring.     (2E51)

Frequently, in fact, it is conversely true that all herrings are red herrings. This often leads to mathematicians speaking of non-red herrings, and sometimes even to a redefinition of herring to include both the red and non-red versions.     (2E52)

[13:23] TerryLongstreth: @Janet ; The first computer bug was a real beetle     (2E53)

[13:24] RaviSharma: excellent summary by Janet     (2E54)

[13:25] RaviSharma: Gary said that you have to be on the label, similar terms, different interpretations, complex relations are possible in ontologies     (2E55)

[13:26] BobbinTeegarden: A triple is just the simplest 'context'     (2E56)

[13:26] TerryLongstreth: @Bobbin - and often the most misleading     (2E57)

[13:30] Robert Rovetto: There's many different forms of definition. Natural language is the most expressive means to express meanings and definitions. Any formal (logical) definition will have expressive limitations. It is the natural language (i.e. human) meaning and understanding that is most critical and should guide formal definitions.     (2E58)

[13:30] Gary Berg-Cross: Seppälä, Selja, Alan Ruttenberg, and Barry Smith. "Guidelines for writing definitions in ontologies." Ciência da Informação 46 (1): 73-88 (2017) Available online at https://philpapers.org/archive/SEPGFW.pdf     (2E59)

[13:30] Seppälä: Guidelines for writing definitions in ontologies: http://revista.ibict.br/ciinf/article/view/4015     (2E60)

[13:30] janet singer: Terry, yes. So many interesting and odd sources of language in STEM     (2E62)

Resources     (2F)

Previous Meetings     (2G)


Next Meetings     (2H)