From OntologPSMW

Jump to: navigation, search
[ ]
Session Planning
Duration 1 hour
Date/Time November 27 2019 17:00 GMT
9:00am PST/12:00pm EST
5:00pm GMT/6:00pm CET
Convener Ravi Sharma


Agenda     (2A)

Conference Call Information     (2B)

Participants     (2C)

Proceedings     (2D)

[12:21] ToddSchneider: From the 2020 Summit page: The theme of the Summit is to examine KGs from a number of points of view ranging from low-level representation and storage techniques to high-level semantics, and from the vendors to the end users. Rather than being split into tracks, the Summit will cover a number of relevant areas with a mix of individual speaker sessions and panel discussion sessions.     (2D1)

[12:23] ToddSchneider: See for a list of 'relevant areas'.     (2D2)

[12:33] ToddSchneider: From the 2020 Summit page:     (2D3)

 Application Domains

[12:34] ToddSchneider: Perhaps use 'Business' instead of 'Financial'; And 'Health Care' instead of 'Medical'.     (2D4)

[12:35] ToddSchneider: We can contact Gary Berg-Cross for examples of specific applications in the Geospatial domain.     (2D5)

[12:36] ToddSchneider: 'Security' or 'Cyber Security' as a domain (of interest).     (2D6)

[12:40] BartGajderowicz: Regarding the "Business" category, would "Economics" be too broad? For example, where would "Insurance" fall under? Would it be "Economics", "Business," or its own category?     (2D7)

[12:47] ToddSchneider: The challenge is to find one or more people to provide presentations in the 'area'.     (2D8)

[12:47] BartGajderowicz: Regarding categories, social domains like cities (e.g. smart city, policy, etc) would be beneficial.     (2D9)

[12:47] John Sowa: We need to distinguish two purposes for any kind of ontology     (2D10)

[12:48] John Sowa: For natural language search, we need immensely large hierarchies     (2D11)

[12:48] John Sowa: WordNet is an example,but it's not big enough.     (2D12)

[12:49] John Sowa: We need a hierarchy of terms that is bigger than the Oxford English Dictionary.     (2D13)

[12:49] John Sowa: In the UMLS web site, they have ontologies with more than 100,000 classes.     (2D14)

[12:51] John Sowa: No ontology with that many classes can have a formal (precise logic) definition of the terms.     (2D15)

[12:52] Janette Wong: There is this paper: Towards Building a Knowledge Base of Monetary Transactions from a News Collection     (2D16)

[12:52] Janette Wong: by Jan R. Benetka et al     (2D17)

[12:53] John Sowa: Cyc is the largest formal ontology on the planet. Doug Lenat stated very clearly that all the precise reasoning is based on the microtheories     (2D18)

[12:57] BartGajderowicz: Followup to my question about social domains, would "Government" be a category? I will try to investigate speakers or examples that can be presented. I know there are many "open gov" portals, some good some bad, but they are out there.     (2D19)

[12:57] RaviSharma1: top level vs microtheories     (2D20)

[12:58] RaviSharma1: todd need to minimize confusion about KGs     (2D21)

[13:03] John Sowa: Fundamental distinction: NL semantics (huge vocabulary, vague definitions)     (2D22)

[13:03] John Sowa: versus formal ontologies: precise definitions for a smaller number of fundamental terms.     (2D23)

[13:07] RaviSharma: John do the terms in formal ontologies relate to KBs data stores and RDBMS etc.?     (2D24)

[13:07] Olaf: Application Domains = Tracks     (2D25)

Prerequisites for Knowledge Graphs
- Distinction: Ontologies/Logic vs Language/Terminologies
- Top level vs. microtheories
- Financial
- Medicine
  - Pharmaceutial
  - Drug Interactions
- Public Health


[13:08] RaviSharma: thanks Olaf     (2D26)

Resources     (2E)

Previous Meetings     (2F)

Next Meetings     (2G)