From OntologPSMW

Jump to: navigation, search
[ ]
Session Krzysztof Janowicz
Duration 1 hour
Date/Time 29 April 2020 16:00 GMT
9:00am PDT/12:00pm EDT
5:00pm BST/6:00pm CEST
Convener KenBaclawski
Track How


Ontology Summit 2020 Krzysztof Janowicz     (2)

Knowledge graphs, closely related to ontologies and semantic networks, have emerged in the last few years to be an important semantic technology and research area. As structured representations of semantic knowledge that are stored in a graph, KGs are lightweight versions of semantic networks that scale to massive datasets such as the entire World Wide Web. Industry has devoted a great deal of effort to the development of knowledge graphs, and they are now critical to the functions of intelligent virtual assistants such as Siri and Alexa. Some of the research communities where KGs are relevant are Ontologies, Big Data, Linked Data, Open Knowledge Network, Artificial Intelligence, Deep Learning, and many others.     (2A)

Agenda     (2B)

  • Krzysztof Janowicz KnowWhereGraph: Enriching and Linking Cross-Domain Knowledge Graphs using Spatially-Explicit AI Technologies to Address Pressing Challenges at the Human-Environment Nexus Slides Video Recording YouTube Video     (2B1)
    • Abstract: Many of the most pressing social and scientific challenges are at the human-environment nexus and thereby at the interface of research fields with vastly different traditions. Therefore, the integration of data remains a prominent research topic. Knowledge graphs and their related technologies are promising candidates for the integration of AI-ready, data-level statements across domains. They provide additional contextual information to researchers, policymakers, nonprofits, and industrial applications. How to extract features from such graphs and integrate them into the downstream models, e.g., in supply chain forecasting or commodity pricing, remains an unsolved problem. In this talk, we will provide an overview of our current ideas and hopes for a KnowWhereGraph that gives broad access to geo-data and spatially-explicit services on top of such data.     (2B1A)
    • Bio: Krzysztof Janowicz is a professor for Geographic Information Science and Geoinformatics at the Geography Department of the University of California, Santa Barbara, USA. He is associate director of the Center for Spatial Studies and former chair of UCSB's cognitive science program. His interests are in geospatial semantics, knowledge graphs, and geographic information retrieval.     (2B1B)
  • Discussion     (2B2)

Conference Call Information     (2C)

Attendees     (2D)

Discussion     (2E)

[11:59] David Eddy: @Janet... hmmmm... "messy language?" What a shock     (2E1)

[12:08] Mark Underwood @knowlengr: Interested in whether any connection/applicability to or These tend to be ArcGIS driven with some unclear NIEM legacy     (2E2)

[12:12] Ken Baclawski: The slides are available at     (2E3)

[12:15] Ravi Sharma: it is working now there was a glitch     (2E4)

[12:26] janet singer: @David, maybe given the interests of current big players messiness and dynamic conditions will finally become the default presumption rather than a surprising deviation from the ideal     (2E5)

[12:26] John Sowa: What is the URL of KJ's slides?     (2E6)

[12:28] Ken Baclawski: @John: The slides are available at     (2E7)

[12:29] ravisharma: Krzysztof - you have shown many thematic maps and some enhancements but because of poor connection due storms in my location in India I have low voice but please explain connection with Knowledge Graphs if any?     (2E8)

[12:32] John Sowa: I noticed one point on KJ's slides: We are not trying to solve the semantic interoperability problem.     (2E9)

[12:33] John Sowa: That is the fundamental problem, and the current ontology standard does not even attempt to solve it.     (2E10)

[12:34] John Sowa: I don't blame KJ for not solving it.     (2E11)

[12:35] ravisharma: Krzysztof - are you saying that combining space imagery with non-space imagery or image Knowledge Graph ready?data such as ground truth or an event that space theme image becomes Knowledge Graph ready? is it processing technology (ArcGIS or ESRI) and added info such as social information that makes the     (2E12)

[12:35] John Sowa: But I do blame the ontology standardizers for suggesting that they are doing anything to address it     (2E13)

[12:41] David Eddy: @John S... does current ontology standard even acknowledge the issue of semantic interoperability EXISTS?     (2E14)

[12:42] ravisharma: Krzysztof - how would you know what is important to add to imagery (thematic) that would make it useful for Knowledge Graph? Phase I.     (2E15)

[12:43] Todd Schneider: David, if you're referring to the new ISO TLO standard, then yes. The need for semantic interoperability is acknowledged.     (2E16)

[12:43] ravisharma: Does a formal language make data graph ready?     (2E17)

[12:44] ravisharma: Krzysztof - how would your COVID-19 graph change if you took out transportation or supply chain?     (2E18)

[12:46] David Eddy: @Todd... the "need"... so is there a specific ISO committee assigned to the task?     (2E19)

[12:47] Todd Schneider: David, I'm not familiar with ISO committees.     (2E20)

[12:47] russell r: Note to self: space [geographic region] is a great integrator in itself     (2E21)

[12:47] John Sowa: Todd, I agree that they throw in some words about it. But Part 1 presents a strategy for specifying independent silos.     (2E22)

[12:47] Ravi Sharma: Krzysztof - KNOW WHERE and Geo Enhancement are value add to geospatial maps for downstream apps?     (2E23)

[12:49] David Eddy: @JohnS / Todd... is there someone in NIST arena looking at semantic interoperability?     (2E24)

[12:53] John Sowa: Google "COBOL Cowboys" for the most fundamental challenge to interoperability: Legacy systems.     (2E25)

[12:53] Todd Schneider: David, I'm not familiar with the work at NIST.     (2E26)

[12:53] David Eddy: @Todd... neither am I     (2E27)

[12:54] Mark Underwood @knowlengr: @David, speaking w/ Leo Obrst about NIST work in this area after Steve Ray left - I think that effort devolved into something less relevant to our community - based on my pretty deep exposure to NIST infosec community     (2E28)

[12:55] Mark Underwood @knowlengr: @KJ would like to see that amazon paper you mentioned     (2E29)

[12:55] David Eddy: @Mark... is Leo still around? Haven't seen anything from him in long time.     (2E30)

[12:56] John Sowa: Current estimate: 80% of the software that is running in applications is written in COBOL. Zero percent of that code uses any formal ontology of any kind. None of the people working on ontology address that issue.     (2E31)

[12:56] Mark Underwood @knowlengr: #Partly retired; I speak to him 1:1 regularly     (2E32)

[12:57] David Eddy: @JohnS... I would seriously question that 80% guess. Last system I worked on 32 years ago (IBM 4300, DOS/VSE) was written in 8 different languages... COBOL being just ONE of them.     (2E33)

[12:58] Mark Underwood @knowlengr: @David You might ask Gary. I ran into him at the NIST lobby quite my accident a couple years back     (2E34)

[12:58] David Eddy: @John S... from seriously old information. Capers Jones offers: 50 major software languages, 1,500 minor languages & a new language is introduced at rate of 1 a month.     (2E35)

[12:58] ravisharma: David Eddy - we had an entire year on this Summit OntologySummit2016 - "Framing the Conversation: Ontologies within Semantic Interoperability Ecosystems" (1G)     (2E36)

[12:59] ravisharma: also 2008 had interoperability track I think.     (2E37)

[12:59] David Eddy: @Ravi... thanks... I'll look back.     (2E38)

[12:59] ravisharma: Todd and i were involved in interoperability track.     (2E39)

[12:59] David Eddy: @Mark... GBC?     (2E40)

[12:59] Evan Wallace: In the Engineering Laboratory (EL) at NIST (where I work) our research is more focused these days on applications to areas where there are specific industry pain points where semantic technology can have nearer term impact.     (2E41)

[13:00] John Sowa: David, that number is not based on the number of applications, but on the number of machine instructions executed per second of all the business and gov't applications around the world.     (2E42)

[13:00] Mark Underwood @knowlengr: Have to bounce to another meeting,..will try to reach KJ directly - Be well, all     (2E43)

[13:01] TerryLongstreth: Transfer from the Zoom Chat: I think this is the Amazon KG entity alignment paper Jano was talking about:     (2E44)

[13:01] Mark Underwood @knowlengr: @Terry, thanks     (2E45)

[13:01] David Eddy: @JohnS... for certain, one of the issues with legacy systems, is we still have no common means of actually measuring what they are. Lines of code is obviously worse than a joke.     (2E46)

[13:02] John Sowa: The major economic systems around the world are locked into software that was written in the 20th c.     (2E47)

[13:03] TerryLongstreth: And I still note with embarrassment that I know Cobol     (2E48)

[13:03] ravisharma: Many thanks to speaker for a great talk need to follow up with phases of your work on KG uses.     (2E49)

[13:05] John Sowa: The number of lines of code for systems (operating systems, TCP/IP transactions, ...), that code is overwhelming more instructions per second than the application code. But today's users, don't write code.     (2E50)

[13:05] BobbinTeegarden: @Krzysztof if you have URLs for other talks you have given please post them for us. Great talk!     (2E51)

[13:07] BobbinTeegarden: @JohnS isn't a map merely a more expressive ontology... ;0)     (2E52)

[13:07] John Sowa: Terry, there is a desperate need for COBOL programmers. You have long-term job security if you know COBOL.     (2E53)

[13:08] Todd Schneider: Meeting ends @13:08 EDT.     (2E54)

Resources     (2F)

Previous Meetings     (2G)

Next Meetings     (2H)

This is a cached copy of the requested page, and may not be up to date.

Sorry! This site is experiencing technical difficulties.

Try waiting a few minutes and reloading.

(Can't contact the database server: Can't connect to local MySQL server through socket '/var/run/mysqld/mysqld.sock' (2) (localhost))

You can try searching via Google in the meantime.
Note that their indexes of our content may be out of date.