Ontolog Forum
OntologySummit2015 Track A Session - Thu 2015-03-12
Topic: : Ontology Integration in the Internet of Things (IoT) [Track A session-2]
Session Co-chairs: RamSriram & LeoObrst
Abstract
Billions of things will be connected to the Internet. These things span a spectrum of cognitive abilities from simple sensors to humans. Ontologies will play a significant role in integrating these things at different abstraction levels. The goal of track A (Ontology Integration in the Internet of Things) is: To discuss the various approaches being taken to address the integration and interoperability issues. We intend to present case studies of IoT, discuss current approaches in integration and interoperability, discuss gaps in current approaches, and discuss issues of vertical integration and interoperability across layers of the IoT, including granularity. We also want to propose methods for achieving integration and interoperability through ontologies, and propose a unified framework for integration and interoperability for multimodal (audio, text, video, etc.) interfaces.
Agenda
- RamSriram (NIST): Toward Internet of Everything: Internet of Things (IoT), Cyber Physical Systems (CPS), and Smart Networked Systems and Societies (SNSS)
- Abstract: The Internet of Everything (IoE) or Smart Networked Systems and Societies (SNSS) will have significant implications for both the market for advanced computing and communication infrastructure and the future markets for nearly 4.5 billion people -- that net-centric societies will create. I will start my presentation with a brief discussion on the evolution of SNSS. The development of a trusted, secure, reliable, and interoperable net-centric computing environment will need technologies that can assure a flexible and scalable system allowing the application of diverse and robust privacy requirements, thus enabling the trusted and meaningful growth of net-centric infrastructures for the benefit of all societies. I will discuss the research challenges (in information technology) for achieving the above net-centric computing environment.
- SpencerBreiner (NIST) and EswaranSubrahmanian (Carnegie Mellon University, NIST): Category Theory for Modular Design: An IoT Example
- Abstract: Category theory (CT) is the mathematical study of structure and, as such, provides a rich formal language for the specification and analysis of ontologies, broadly construed as structured representations of information. In this talk we will use constructions from CT to define a formal model for a smart heating system in a house. Although simplistic, the model highlights various advantages of the category theoretic approach including modularity, extensibility, rich semantics and deep connections with the theory of databases and programming languages. As an integrative formalism, Category theoretical models could also serve in the requirements definition of IoT components and as a substrate for learning algorithms using the data base to evolve the underlying behavioral models of the components.
- KrzysztofJanowicz (University of California, Santa Barbara): Ontology Virtualization For Smart Environments
- Abstract: The Internet of Things is more than just connected devices. It is about the interaction of cyber-systems with the physical environment, individuals, and society. In such a heterogeneous setting where multi-thematic, multi-perspective, multi-media data have to be handled at different spatial, temporal, and thematic granularities, semantic interoperability becomes a key challenge. While ontologies and Semantic Web technologies have been proposed to address data integration and interoperability, monolithic ontologies do not perform well in highly heterogeneous settings. Ontology design patterns have been proposed to handle this issue by providing modular but self-contained building blocks. Additionally, new research shows promise in approaching the knowledge engineering bottleneck by means of ontology learning. While both approaches aim at scaling ontology engineering, it is also worth exploring how to scale the (re)use of ontologies. In this talk we introduce the notion of ontology virtualization, namely to abstracts from the underlying patterns and axiomatization to provide flexible plug&play-style reconfiguration of patterns together with purpose-driven 'semantic views'.
Prepared presentation material
Prepared material (slides) from the presenters can be accessed by clicking on each of these links: Chair Intro ; RamSriram ; SpencerBreiner-EswaranSubrahmanian ; KrzysztofJanowicz .
Audio Recording of the Session
In-Session Chat Transcript
[12:22] Leo Obrst: Ontology Summit 2015 Track A Session - Thu 2015-03-12 Ontology Integration in the Internet of Things (IoT) [Track A session-2]
Agenda
Ram D. Sriram (NIST) and Leo Obrst (MITRE): Track A Co-Champions: Introduction to the Session
Ram D. Sriram (NIST): Toward Internet of Everything: Internet of Things (IoT), Cyber Physical Systems (CPS), and Smart Networked Systems and Societies (SNSS)
Spencer Breiner (NIST) and Eswaran Subrahmanian (Carnegie Mellon University, NIST): Category Theory for Modular Design: An IoT Example
Krzysztof Janowicz (University of California, Santa Barbara): Ontology Virtualization for Smart Environments
Discussion
Slides: http://ontolog-02.cim3.net/w/index.php?title=ConferenceCall_2015_03_12&oldid=15577#hid1E
[12:26] anonymous morphed into Al Jones
[12:26] Peter P. Yim: Hello Al ... haven't seen you for while!
[12:27] Leo Obrst: I've logged in as admin on the bridge.
[12:28] Peter P. Yim: a better link to the slides: http://ontolog-02.cim3.net/wiki/ConferenceCall_2015_03_12#Prepared_presentation_material ... (this links to the latest version of the page content)
[12:31] Ravi Sharma: Peter - How are you? the link brings back to the conf. session page, but can we still open the presentation screen, I download the papers and follow the speakers by slide numbers.
[12:34] Padi Sarala: Hello Everybody
[12:34] Mark Underwood: FYI I'm having trouble connecting to Skype
[12:34] Peter P. Yim: @RaviSharma - we don't have shared-screen support any more, but downloading the slides and following the speakers' slide number prompts
[12:35] Peter P. Yim: ... would be the way
[12:42] Peter P. Yim: @[9:34] Mark Underwood - are you in yet; try dialing the phone number too, if skype doesn't work for you (it does for me) - ref. http://ontolog-02.cim3.net/wiki/ConferenceCall_2015_03_12#Conference_Call_Details
[12:43] Mark Underwood: @Peter - Thx, connected after a couple kills [12:43] Krzysztof Janowicz: Is there a VNC Link?
[12:43] Leo Obrst: There is no more VNC link.
[12:43] Krzysztof Janowicz: Thanks.
[12:43] Peter P. Yim: @KrzysztofJanowicz - No, no vnc support any more
[12:44] Dennis Wisnosky: The slides are not downloading for me. I could see Leo's slides, not these.
[12:44] Krzysztof Janowicz: They are large, you may have to wait a few minutes
[12:44] Leo Obrst: Slide 13
[12:50] Dennis Wisnosky: took 10 minutes.
[12:55] Todd Schneider: Ram, Where do you see foundational ontologies (e.g., BFO, DOLCE, etc.) coming into play?
[12:58] JackRing1: slide 51. There may be better ways than 'test'
[13:02] Ravi Sharma: @Sriram - hence slides for integration of Social-spatio-temporal image integration, for operational systems such as DHS, CDC, Financial flows, etc. you need different sub-settings of ubiquitous integrated IoT and images, any ideas on what metadata standards will evolve that can meet such future scenarios that you have painted?
[13:02] Mark Underwood: FYI our session hash tag is #ontologysummit - ripe for sentiment analysis
[13:03] JackRing1: What if some of the Things in the IoT are a million software bugs?
[13:04] JackRing1: slide 65 also mood sensing
[13:05] Steve Ray: @Jack: I think that biomimetics is the best blueprint for these kinds of emergent systems. That is, we'll have viruses, antibodies, diseases, etc.
[13:06] JackRing1: 70 interop with schemas and data dictionaries.
[13:10] Peter P. Yim: great talk, Ram, thank you!
[13:10] JackRing1: Steve Ray, yes we will have those but how to mitigate?
[13:11] Ravi Sharma: @Sriram - you have answered partly but still sub-setting dynamically and metadata selection - such as adaptive algorithms are still an open Q if you like to address?
[13:11] Steve Ray: @Jack: We will need to build the antibodies and other disease-fighting systems!
[13:12] John Morris: BTW, the Twitter handle for NIST is "USNISTgov", not "NISTgov", as far as I can tell.
[13:14] Ram D. Sriram: Lots of great questions.
[13:17] Ram D. Sriram: @JaackRing: Software assurance is one of the challenges I did not dwell on in more detail
[13:18] Mark Underwood: @JohnM - thx for the correction. Oy.
[13:20] Dennis Wisnosky: Slide 6 needs and AND gate
[13:24] John Morris: This is Spencer Breiner speaking now, yes?
[13:24] Mark Underwood: Reference for the connection between category models & DB schemas?
[13:26] Ravi Sharma: There is a noise on the audio that is generated by being close to a device such as phone, fan, or something like that, please turn it away
[13:31] Ram D. Sriram: @Ravi: Spencer is using a speaker phone
[13:32] Leo Obrst: There are some folks in a common room.
[13:34] Conrad Beaulieu: How is category theory formalized in a data specification & functional language?
[13:35] Peter P. Yim: @SpencerBreiner & Eswaran Subrahmanian - are you tracking/collaborating with the OntoIOp standardization effort (ref. http://ontoiop.org ), where a few of the folks on this call are involved - in that work, category theoretic semantics (among others) have been applied
[13:35] Peter P. Yim: ... [ OntoIOp Semantics: direct set-theoretical (+ institutional, category-theoretical); translational; "collapsed" ( source: http://www.cicm-conference.org/2012/posters/ChristophLange-DistributedOntologyLanguage.pdf ) ]
[13:36] Steve Ray: Interesting slide numbering technique.
[13:37] Mark Underwood: @SteveRay Adobe had to adopt a dual numbering scheme - sobering
[13:38] Conrad Beaulieu: @PeterYin: thank you for the reference!
[13:41] Fabian Neuhaus: @ConradBeaulieu: the most recent version of the DOL draft specification is available here https://github.com/tillmo/DOL/raw/master/Standard/ebnf-OMG_OntoIOp_current.pdf
[13:45] Tara Athan: Category theory is not just in Haskell - here is a promising library in Scala https://github.com/non/cats
[13:46] Mark Underwood: Reference for Breiner's "Category Web Project" ?
[13:48] Al Jones: This is Spencer & Sub, answering questions from Al's computer.
[13:48] Steve Ray: @Mark: The link was invisible, but it was there: http://categoricaldata.net/
[13:49] Tara Athan: In environmental economics, there is also a notion of consumable versus non-consumable resources.
[13:49] Leo Obrst: Mouse over the "Categorical data project" bullet on Spencer's last slide to see the url, connect.
[13:50] Jack Ring: Ram. When may we dwell on software assurance? The presence of logic, arithmetic and semantic faults in the code establish the vulnerability of the system.
[13:50] Peter P. Yim: @FabianNeuhaus - would you share a link to an (more recent) overview of the OntoIOp / DOL effort, please (there must have been slidedecks available online, I wasn't able to locate them that would give people a good intro)
[13:50] Al Jones: @Conrad, are you asking about how one formalizes category theory (e.g., in another logical formalism)? If so, one only requires a fragment of first-order logic including conjunction and equality.
[13:52] Al Jones: @Peter, I (Spencer) am pretty new to this area, but Fabian pointed us to DOL a few weeks ago. We look forward to working with their group, but we haven't really started yet.
[13:53] Peter P. Yim: Thank you, Spencer
[13:53] Al Jones: @Tara, Thanks for the pointer! It's edifying that the more people we ask the more applications that we find already in the wild.
[13:55] Ram D. Sriram: @JackRIng: That is a problem we are planning on addressing for mPCDs (Paul Black, paul.black@nist.gov is the POC here)
[13:56] Al Jones: @Mark, you can find a discussion of the relationship between category theory and databases in the FQL section of categoricaldata.net.
[13:56] Jack Ring: Steve Ray: Don't need antibodies if the system is impervious to the capabilities of the attacker.
[14:01] Anatoly Levenchuk: INCOSE Russian chapter have a research topic about category theory as an formalism for model-based systems engineering. Leader of this is Dr.Sergey Kovalyov. We also know other category theory proponents that think about ontology of a System and systems engineering applications (e.g. Kenneth A.Lloyd).
[14:03] Terry Longstreth: @KrzysztofJanowicz Slied 6 - Do your upper layers promote the creation of ontology evolving software?
[14:03] Bobbin Teegarden: @Krzysztof Don't the Mappings turn into a sort of upper-level (if light weight) ontology?
[14:04] Steve Ray: @Jack: That sounds like a firewall - don't you think it is better to build immunity into the components of a system rather than trying to isolate the system inside a bubble?
[14:06] Peter P. Yim: @KrzysztofJanowicz - re your slide 5 assertion that "Monolithic (upper-level) ontologies do not perform well in highly heterogeneous settings" ... I think "upper level" seems misplaced ... would "(Very granular) Monolithic ontologies do not perform well in highly heterogeneous settings" reflect better what you meant to say?
[14:06] John Graybeal: On Janowicz slide 8, oddly the Source of a Fix is subClassOf ssn:Device. Couldn't a person also be a Source?
[14:08] John Graybeal: @PeterYim: My intuition and experience has been that working with upper-level ontologies that try to cover a lot of areas is problematic, because they have so many broad assumptions that you run into trouble somewhere in your more precise local system. Not sure if that's what he's saying, but that's what resonated for me.
[14:09] Peter P. Yim: ... personally, I believe there is a huge need for the shared-adoption or standardization of upper-level ontologies (just imagine we have a million different notions of "Time")
[14:10] Mark Underwood: Interesting concept - analogy / design pattern of virtualization applied to ontologies
[14:11] Mike Bennett: @PeterYim xx:06 I'm inclined to agree with you. One possible source of confusion is whether what one means by an "ontology" is itself highly axiomatized, or is a simple generic ODP such as the Trajectories one - having this kind of thing in upper / mid level ontologies should assist integration; trying to say more about each thing might not.
[14:11] Peter P. Yim: @JohnGraybeal: agree, and hence I qualified that statement with "Very granular" in "(Very granular) Monolithic ontologies do not perform well ..."
[14:11] uri shani: which slide, which talk?
[14:11] Ravi Sharma: @Krzysztof - Can you explain a bit more on why provenance is not also related to time evolution of observation?
[14:12] Peter P. Yim: @uri shani - Krzysztof Janowicz speaking just now (just finished ...)
[14:13] John Graybeal: @PeterYim, yes ok re 'granular', interestingly I understood 'monolithic' to mean to cover everything, hence it would have to be granular.
[14:13] Krzysztof Janowicz: John Graybeal: yes, thanks for the question and sorry for not making this clear. the dashed lines are semantic hooks to other patterns
[14:13] Krzysztof Janowicz: the trajectory pattern's source class can be used for humans, devices, and even simulations
[14:14] PatrickOBrien1: For Krzysztof Janowicz slide 7. Can he provde an example of "Some patterns are strategies"?
[14:17] Steve Ray: @Krzysztof: Each of your design patterns has an implicit context that bounds its applicability. Do you have some method of capturing that information? i.e. metadata for the pattern?
[14:17] Peter P. Yim: @JohnGraybeal - I took 'monolithic' to mean "not modular" or "not layered" ... By virtual of "upper level" (which probably implies that there are other levels (like the generally accepted notions of a "middle level ontologies," "super-domain ontologies," "domain ontologies," etc.) which then makes "upper level" in the original assertion dubious
[14:18] JackRing1: Steve Ray: exactly. That is why we must cleanse all software of logic, arithmetic and semantic faults and do so as often as the software is changed. This essential aspect is being vigorously ignored by IoT champions. Some even say it can't be done.
[14:19] Krzysztof Janowicz: PatrickOBrien1: Yes. This is similar to software engineering design patterns. A strategy would be using concept reification for time-indexed relations (just to give one example)
[14:19] PatrickOBrien1: Thank you [14:19] Krzysztof Janowicz: "'monolithic' to mean "not modular" or "not layered"" --> yes
[14:20] Victor Agroskin: @Krzysztof Ontology patterns are standardised now by the W3C RDF Data Shapes Working Group http://www.w3.org/2014/data-shapes/wiki/Main_Page . Do you see value in an ontology pattern language they are proposing?
[14:21] JackRing1: Is anyone working on an ontology regarding Do No Harm?
[14:21] Ravi Sharma: @ all - thanks
[14:21] Krzysztof Janowicz: "upper level" --> here just as shortcut for ontologies that introduce strong ontological commitments above the layer of concrete purpose-driven cases.
[14:22] Krzysztof Janowicz: one example would be a domain ontology for geographic features such as 'forest'
[14:24] Krzysztof Janowicz: Bobbin Teegarden: interesting point, thanks. The big difference is that this remains configurable in a plug&play manner without the need to arrive at 'agreement' first.
[14:24] Ravi Sharma: static objects and subjects, ok but forest that is living has lot more?
[14:24] Krzysztof Janowicz: IMHO, ontologies should make differences explicit not try to arrive at some sort of common truth.
[14:25] JackRing1: Can't get there. One way is to apply Category Theory to discover the ontology of "ontology"
[14:25] John Morris: A really great session!
[14:25] Victor Agroskin: A year ago at Summit Hakathon http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2014_Hackathon_ReferenceDataForAnimeAndManga we've experimented with patterns to describe schema-neutral representation of data for several datasources. This work was based on ontology templates introduced by ISO 15926-7 which is an attempt to bring ontology models into engineering domain.
[14:25] Krzysztof Janowicz: Ravi Sharma: it is. Observations need provenance data, namely the used observation procedures, sensors, etc.
[14:26] Anatoly Levenchuk: @JackRing We here thinking about category theory as an ontology language similar way
[14:26] Krzysztof Janowicz: Victor Agroskin: Thank you very much for the pointer!
[14:27] Krzysztof Janowicz: Terry Longstreth: interesting question, have not thought about it this way.
[14:27] Ravi Sharma: @Ram - tremendous work putting it together
[14:28] JackRing1: @AnatolyLevenchuk: yes. leads us to see types of systems as well as classes of systems. Will send you a slide.
[14:28] Ravi Sharma: thanks to Leo and Ram
[14:28] Mark Underwood: Janowicz: +1 for the design pattern approach overall. Next step: routinized integration w/ Visual Studio
[14:28] Anatoly Levenchuk: Very, very, very interesting session!
[14:28] JackRing1: +1
[14:28] Peter P. Yim: awesome session; great talks! ... Thank you to co-chairs and Ontology Summit organizing committee folks!
[14:28] Leo Obrst: Thanks, all!
[14:28] Mark Underwood: High quality as always - thanks all
[14:28] Krzysztof Janowicz: Steve Ray: excellent point, I will try to answer this after the session.
[14:29] Krzysztof Janowicz: Thanks to Leo,Ram
[14:29] Steve Ray: @Krzysztof: Thanks, I'm still online.
[14:31] Ravi Sharma: Also All speakers thanks.
Additional Resources
...
Conference Call Details
- Date: Thursday, 12-Mar-2015
- Start Time: 9:30am PDT / 12:30pm EDT / 5:30pm CET / 4:30pm GMT / 1630 UTC
- ref: World Clock
- Expected Call Duration: ~2 hours
- Dial-in:
- Phone (US): +1 (425) 440-5100 ... (long distance cost may apply)
- ... [ backup nbr: (315) 401-3279 ]
- when prompted enter Conference ID: 843758#
- Skype: join.conference (i.e. make a skype call to the contact with skypeID="join.conference") ... (generally free-of-charge, when connecting from your computer ... ref.)
- when prompted enter Conference ID: 843758#
- Unfamiliar with how to do this on Skype? ...
- Add the contact "join.conference" to your skype contact list first. To participate in the teleconference, make a skype call to "join.conference", then open the dial pad (see platform-specific instructions below) and enter the Conference ID: 843758# when prompted.
- Can't find Skype Dial pad? ...
- for Windows Skype users: Can't find Skype Dial pad? ... it's under the "Call" dropdown menu as "Show Dial pad"
- for Linux Skype users: please note that the dial-pad is only available on v4.1 (or later; or on the earlier Skype versions 2.x,) if the dialpad button is not shown in the call window you need to press the "d" hotkey to enable it. ... (ref.)
- Phone (US): +1 (425) 440-5100 ... (long distance cost may apply)
- In-session chat-room url: http://webconf.soaphub.org/conf/room/summit_20150312
- instructions: once you got access to the page, click on the "settings" button, and identify yourself (by modifying the Name field from "anonymous" to your real name, like "JaneDoe").
- You can indicate that you want to ask a question verbally by clicking on the "hand" button, and wait for the moderator to call on you; or, type and send your question into the chat window at the bottom of the screen.
- thanks to the soaphub.org folks, one can now use a jabber/xmpp client (e.g. gtalk) to join this chatroom. Just add the room as a buddy - (in our case here) summit_20150312@soaphub.org ... Handy for mobile devices!
- Discussions and Q & A:
- Nominally, when a presentation is in progress, the moderator will mute everyone, except for the speaker.
- To un-mute, press "*7" ... To mute, press "*6" (please mute your phone, especially if you are in a noisy surrounding, or if you are introducing noise, echoes, etc. into the conference line.)
- we will usually save all questions and discussions till after all presentations are through. You are encouraged to jot down questions onto the chat-area in the mean time (that way, they get documented; and you might even get some answers in the interim, through the chat.)
- During the Q&A / discussion segment (when everyone is muted), If you want to speak or have questions or remarks to make, please raise your hand (virtually) by clicking on the "hand button" (lower right) on the chat session page. You may speak when acknowledged by the session moderator (again, press "*7" on your phone to un-mute). Test your voice and introduce yourself first before proceeding with your remarks, please. (Please remember to click on the "hand button" again (to lower your hand) and press "*6" on your phone to mute yourself after you are done speaking.)
- Please review our Virtual Session Tips and Ground Rules - see: VirtualSpeakerSessionTips
- RSVP to ram.sriram@nist.gov with your affiliation appreciated, ... or simply just by adding yourself to the "Expected Attendee" list below (if you are a member of the community already.)
- This session, like all other Ontolog events, is open to the public. Information relating to this session is shared on this wiki page: http://ontolog-02.cim3.net/wiki/ConferenceCall_2015_03_12
- Please note that this session may be recorded, and if so, the audio archive is expected to be made available as open content, along with the proceedings of the call to our community membership and the public at-large under our prevailing open IPR policy.
Attendees
Al Jones, Allyson, Anatoly Levenchuk, Antoine Gerardin, Bobbin Teegarden, Christopher Spottiswoode, Conrad Beaulieu, Dennis Wisnosky, Fabian Neuhaus, Gary Berg-Cross, Jack Ring, Joe Kopena, Joel Bender, John Graybeal, John Morris, Josh Lieberman, Judith Gelernter, Krzysztof Janowicz, Leo Obrst, Liana Kiff, Malek, Mark Underwood, Matthew West, Michael Grüninger, Mike Bobak, Mike Bennett, MikeBobak, Osman Bineev, Padi Sarala, PatrickOBrien, Peter P. Yim, Ram D. Sriram, Ravi Sharma, Richard Martin, Rokan, Steve Ray, Tara Athan, Terry Longstreth, Todd Schneider, Tom Tinsley, Veda Storey, Victor Agroskin, uri shani