ppy/OntologySummit2014-s03_chat-transcript_edited_20140130b.txt ------ Chat transcript from room: summit_20140130 2014-01-30 GMT-08:00 [PST] ------ [8:50] PeterYim: Welcome to the = OntologySummit2014 session-03 Track-B: Making use of Ontologies: Tools, Services, and Techniques - I - Thu 2014-01-30 = Summit Theme: Summit Theme: OntologySummit2014: "Big Data and Semantic Web Meet Applied Ontology" Session Topic: Track-B: Making use of Ontologies: Tools, Services, and Techniques - I Session Co-chairs: Dr. ChristophLange (University of Bonn / Fraunhofer IAIS), Professor AlanRector (University of Manchester) Briefings: * Professor TillMossakowski (University of Magdeburg) - "Challenges in Scaling Tools for Ontologies to the Semantic Web: Experiences with Hets and OntoHub" * Dr. ChrisWelty (IBM Research) - "Semantic Technology in Watson" * Professor AlanRector (University of Manchester) - "Axioms & Templates: Distinctions & Transformations amongst Ontologies, Frames, & Information Models - or - OWL, UML, and Frames" Logistics: * Refer to details on session page at: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2014_01_30 * (if you haven't already done so) please click on "settings" (top center) and morph from "anonymous" to your RealName * Mute control (phone keypad): *7 to un-mute ... *6 to mute * Attn: Skype users ... see: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2014_01_30#nid44MF ** you may connect to (the skypeID) "joinconference" whether or not it indicates that it is online (i.e. even if it says it is "offline," you should still be able to connect to it.) ** if you are using skype and the connection to "joinconference" is not holding up, try using (your favorite POTS or VoIP line, etc.) either your phone, skype-out or google-voice and call the US dial-in number: +1 (206) 402-0100 ... when prompted enter Conference ID: 141184# ** 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: if the dialpad button is not shown in the call window you need to press the "d" hotkey to enable it * when posting in this Chat-room, kindly observe the following ... ** whenever a name is used, please use the full WikiWord name format (every time you don't, some volunteer will have to make an edit afterwards) ** always provide context (like: "[ref. JaneDoe's slide#12], I think the point about context is great" ... rather than "that's great!" as the latter would mean very little in the archives.) ** when responding to a specific individual's earlier remarks, please cite his/her full WikiWord names *and* the time-stamp (in PST) of his/her post that you are responding to (e.g. "@JaneDoe [11:09] - I agree, but, ...") ** use fully qualified url's (include http:// ) without symbols (like punctuations or parentheses, etc. right next to it) Attendees: AbhayKashyap, AlanRector, AleksandraSojic, AliHashemi, AmandaVizedom, AnatolyLevenchuk, AndreaWesterinen, AndreaWesterinen, BartGajderowicz, CarmenChui, ChrisWelty, ChristineKapp, ChristophLange, ConradBeaulieu, DanielMcShan, DennisWisnosky, DennisPierson, EarlGlynn, EarlGlynn, EdBernot, ElizabethFlorescu, EricChan, FranLightsom, GarrettClarke, GaryBergCross, GenZou, HaroldBoley, HensonGraves, JamesOverton, JeffCox, JensOrtmann, JohnMcClure, JulienCorman, KenBaclawski, LamarHenderson, LeoObrst, LesMorgan, LianaKiff, MarciaZeng, MariaPoveda, MarkFox, MatthewLange, MeganKatsumi, MichaelGruninger, MikeBennett, MikeDean, NancyWiegand, OliverKutz, OndrejZamazal, PeterMidford, PeterYim, RamSriram, SimonSpero, SundayOjo, TaraAthan, TerryLongstreth, TillMossakowski, TimDarr, TimFinin, ToddSchneider, TorstenHahmann, VictorAgroskin, == Proceedings == [8:59] anonymous morphed into MarciaZeng [9:18] anonymous morphed into ChrisWelty (1) [9:18] Chris Welty (1) morphed into ChrisWelty [9:22] ChrisWelty: hello [9:23] PeterYim: Hi Chris, Hi Alan, Hi Oliver [9:23] ChrisWelty: hi peter [9:23] ChrisWelty: anything i need to do? [9:25] ChrisWelty: is it preferable for me to join by skype or phone? [9:27] PeterYim: Yes, Chris ... please use a land-line if you can [9:28] GaryBergCross: Settings can also be used to show timestamps so we can indicate what time the comment we are responding to was made. It helps tracking conversation threads. [9:29] anonymous1 morphed into LesMorgan [9:30] anonymous morphed into CarmenChui [9:30] anonymous morphed into JeffCox [9:30] AmandaVizedom morphed into AmandaVizedom [9:30] Alan Rector (2) morphed into AlanRector [9:32] anonymous1 morphed into EarlGlynn [9:32] anonymous morphed into MarkFox [9:32] anonymous morphed into JulienCorman - [9:32] AmandaVizedom: Peter, your instructions are currently very hard to hear over background white-noise - [9:33] AmandaVizedom: white-noise now gone [9:34] anonymous1 morphed into PEM [9:34] PeterYim: == ChristophLange starts the session on behalf of the Track-B session co-champions ... see slides under: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2014_01_30#nid4564 [9:37] anonymous morphed into AleksandraSojic [9:38] anonymous1 morphed into OndrejZamazal [9:38] anonymous2 morphed into TimDarr [9:39] anonymous morphed into ToddSchneider [9:40] anonymous morphed into ConradBeaulieu [9:40] PEM morphed into PeterMidford - [9:41] PeterYim: @PEM ... kindly morph into your real name (in WikiWord format, preferably) for attribution purposes - [9:41] PeterYim: @PeterMidford ... thank you (you beat me to it!) [9:41] PeterYim: == TillMossakowski presenting ... [9:43] anonymous morphed into DennisPierson [9:43] PeterYim: @Christoph ... there are about 10 people on the conference bridge (voice) who are not in the chat-room yet ... in between speakers, please prompt them to join us in the chat-room too (see details at the top of the session page) [9:44] ChristophLange: @PeterYim: OK, will do ASAP [9:43] anonymous morphed into LamarHenderson - [9:47] PeterYim: ... on slide#11 now [9:47] ChristophLange: ref. AlanBundy, the ontology researcher who TillMossakowski mentioned on slide#9 - see: http://homepages.inf.ed.ac.uk/bundy/ [9:48] anonymous morphed into TorstenHahmann [9:56] anonymous morphed into AbhayKashyap [9:49] ChristophLange: GitHub as mentioned by TillMossakowski on slide#12: http://github.com - [9:52] AmandaVizedom: on slide 14 now [9:53] JensOrtmann: what does MMT stand for? [9:54] ChristophLange: re:MMT (slide#14) Module System for Mathematical Theories: https://svn.kwarc.info/repos/MMT/doc/html/index.html [9:54] JensOrtmann: ok great thank you! [9:55] TillMossakowski: (slide#15) correction, url should be: http://ontohub.org/colore/algebra/vectorspace.clif [9:57] TaraAthan: The link http://ontohub.org/colore/algebra/vectorspace.xml provides metadata for the CLIF ontology in an XML format, not an XML-serialization of the axioms. [9:58] anonymous morphed into MatthewLange [9:59] ChristophLange: slide#16: OOPS Ontology Pitfall Scanner ( http://oeg-lia3.dia.fi.upm.es/webOOPS/catalogue.jsp ) [10:00] MariaPoveda: or http://www.oeg-upm.net/oops/ [10:00] MariaPoveda: you can use http://www.oeg-upm.net/oops/ that is independent of the server it is installed in [9:59] MariaPoveda: :-/ please let me know every time you have problems with OOPS! services [10:00] ChristophLange: In fact MariaPoveda will present in our 13 March session [9:59] GaryBergCross: @Till Great connection to the semantic content reuse topic of Track A. [10:03] GaryBergCross: @Till when you get a chance can you provide some more references/links to the tools and technology you mentioned. We can eventually get these into our library too... [10:03] TillMossakowski: http://hets.dfki.de, http://ontohub.org [10:01] anonymous morphed into GarrettClarke [10:02] anonymous morphed into LamarHenderson [10:04] RamSriram: Worth looking at http://asterixdb.ics.uci.edu/ for "BIG ONTOLOGIES" [10:05] MariaPoveda: @TillMossakowski do you know http://neon-toolkit.org/wiki/KC-Viz ? it is for getting the key concepts from an ontology, it might be a first step for the summarization [10:05] MariaPoveda: sorry, for splitting [10:05] TillMossakowski: thanks, I do not know it, will have a look [10:06] MariaPoveda: there are also plugins for modularization http://neon-toolkit.org/wiki/Ontology_Module_Extraction and http://neon-toolkit.org/wiki/Ontology_Module_Partition [10:07] ChristophLange: @MariaPoveda: I assume that most of these modules are specific to OWL, right? [10:11] MariaPoveda: right [10:13] AmandaVizedom: @MariaPoveda, am I right in recalling that the NeOn work is specifically for DL? Or has it expanded? [10:16] MariaPoveda: @AmandaVizedom I'm not 100% sure ( :( ) but I think you are right [10:16] TillMossakowski: I think that NeOn also covers F-logic. [10:21] MariaPoveda: sorry, I can't find any information about it, my experience with NeOn was only about OWL [10:24] AlanRector: Actually NeOn is at least as much about F-Logic as OWL, and supports a number of different paradigms. My link is EnricoMotta at the Open University for details e.motta [at] open.ac.uk [10:03] PeterYim: == ChrisWelty presenting ... [10:28] ChristophLange: important background info by ChrisWelty about slide#17: the average named entity refers to 5 distinct things [10:28] AlanRector: For modularization, see also the modularization tools at owl.cs.manchester.ac.uk. There is a whole subfield on how to break OWL ontologies down into modules, where "module" is used explicitly as meaning a subontology for a "signature" - i.e. a set of entity symbols - such that all inferences amongst the members of the signature that would be made in the entire ontology will be made in the sub-ontology/module. [10:30] AmandaVizedom: for @ChrisWelty: [10:31] ChristineKapp: @ChrisWelty: I'm sure it was just white font [10:31] PeterYim: re: slide#25 not showing up properly in the pdf version ... I have now posted ChrisWelty's original version in ppsx - see: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2014_01_30#nid4564 [11:16] PeterYim: @ChrisWelty [10:30] and All, I have made an attempt to correct the problematic pdf conversion of your (Welty) slides, and have swapped that in ... slides 17, 22 & 25 now shows the correct image in the [2-Welty] slide deck now at: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2014_01_30#nid4564 [10:33] AmandaVizedom: "Part of system is understanding just how complete or incomplete a source is when it comes to providing a certain type of evidence" - @ChrisWelty, discussing slide 26. I appreciate this meta-knowledge use; it is so critical for useful info systems of many types. - [10:33] ChrisWelty stepping out for 2 mins for a bio-break - [10:40] ChrisWelty is back [10:40] GenZou: @ChrisWelty: What do you think are the advantages and disadvantages of converting natural language queries to formal queries and answer them by reasoning? Why did not Watson also use it as one option to get answers? [10:42] ChrisWelty: converting nlp to logic is impossible to do reliably, and the knowledge needed to answer is not represented in logic, its in text [10:49] GenZou: @ChrisWelty: IC. Does Watson use any other kinds of ontological reasoning besides spatial reasoning and temporal reasoning to score the answers or for other tasks? [10:33] PeterYim: == AlanRector presenting ... [10:37] ChristophLange: AlanRector said about slide#3 that he works with "medium" rather than "big" data (ontologies up to 1 million classes) [10:38] MatthewLange: Does anyone have a reference to the ICD-9/10 and modifications, which outlines the changes @AlanRector is speaking of? [10:44] MariaPoveda: @MatthewLange I was not here when he referred to ICD-9/10, could be this http://www.who.int/classifications/icd/en/ ? [10:44] TimFinin: +1 for Ontology != KB [10:45] anonymous1 morphed into BartGajderowicz [10:46] ChristophLange: @MatthewLange, @MariaPoveda: I recall there were some publications on formalising the ICD in OWL; was it this one? Manuel Moeller, Michael Sintek, Ralf Biedert, Patrick Ernst, Andreas Dengel, Daniel Sonntag: Representing the International Classification of Diseases Version 10 in OWL. KEOD 2010: 50-59 ( http://www.dfki.de/~sonntag/I3CK-selection-2012.pdf ) [11:03] MatthewLange: @ChristophLange thanks for the link, but I was referring not to formalization of ICD into OWL, but to the United states specific modifications to ICD that cast a wider, and deeper net for US-specific things (like the NASA astronaut example...) @AlanRector, could you please provide a reference when you are done speaking? [10:47] AmandaVizedom: Slide #7 seems to me to be a red herring. I would argue that one of those is a confusion between mental objects and formal models, the other is a confusion between the reality and the model. The latter is slightly mitigated by accepting the idea of contextualization by application requirements, but that is not the "original" philosophical meaning (it's closer to the idea of a scientific model in philosophy of science). [10:47] TillMossakowski: Isn't "Pneumonia may be caused by bacteria" also universal knowledge, but the point is that it cannot be axiomatized in OWL, but needs a modal logic? [10:53] AmandaVizedom: @AlanRector, slide #10: at least some of this seems to be about, or dependent on, particular representation choices (for example, the idea that template statements involve primitive concepts only), even within the range of ontology languages. Your intro to it, however, gave me the impression that you were trying to say something about types of information represented. Can you explain? - [10:51] ChristophLange: slide#11 [10:56] AmandaVizedom: AlanRector, slide 11: quite true, though it's also true that in many operational spaces, information artifacts are some of the things in the domain, and there may be a need to represent our understanding of both some kinds of information artifacts and the things those information artifacts are about. Problems come, I think, when people aren't really aware of which they are doing when. [11:01] AndreaWesterinen: Slide #14, Although non-standard, there are reasoners/stores that do integrity constraint checking via axioms (e.g., Stardog). This distinction was valuable. [11:02] AmandaVizedom: @AlanRector slide 14: This point about OWL domain and range constraints not really being constraints (but rather axioms from which inferences can be drawn) is a such an important point, and so insufficiently understood by many doing OWL modeling. IME (in my experience), many people who do appreciate it end up putting their constraints into SPARQL queries, losing any declarative representation of constraint relationship. Is that also what you see? [11:05] AmandaVizedom: @AlanRector, when done speaking, will you provide a reference for the Motik paper? - [11:06] ChristophLange: slide#20 - [11:10] ChristophLange: slide#23 [11:11] anonymous morphed into LamarHenderson [11:12] ChrisWelty: I have a hard stop at 14:30 EST (in ~20 mins) [11:14] GaryBergCross: @Alan, when you get a chance could you say a bit more about the tools you use or are interested in? [11:18] AlanRector: @GaryBergCross - as a quick caricature, the sort of tools I'd like to see is a variant of Protege or the Neon or of a UML editor that expressed things in ways familiar to users but then translated them into something like the representation in my presentation - typically generating half-a-dozen axioms for each visible statement as seen by the user. [11:19] GaryBergCross: Alan Thank you. Do you think that controlled natural language tools can play a role in getting started on KBs? [11:14] PeterYim: == Q & A and Open Discussion ... [11:14] PeterYim: ChristophLange - our focus: OntologySummit2014: "Big Data and Semantic Web Meet Applied Ontology" - Track-B: Making use of Ontologies: Tools, Services, and Techniques [11:15] ChristophLange: "big picture question": AlanRector asked the question whether "ontology" is the same as "knowledge representation". Now from Watson it seems that offering knowledge-rich services doesn't need ontologies or ontology languages in a traditional sense. On the other hand there is a lot of work on improving ontology languages, as we heard from TillMossakowski, and there are better ways of using OWL, as we heard from Alan. So let me ask: On the Web of Data, are ontologies and ontology languages, and tools and techniques for them, are they still relevant? [11:15] ChrisWelty: was the question directed at a particular person? [11:15] MikeBennett: Surely "ontology" is a heteronym - it means both of the things stated (a formal, decidable resource, and a formal representation of some business concepts). [11:16] AmandaVizedom: @AlanRector: re slide 23 and compositionality - to me, this is one issue that makes me miss having functions (including those that result in a denotational term with a defined relationship to their arguments) in an ontology language. Thoughts? [11:17] GaryBergCross: I took part of Alan's presentation to be (in a sense) that things like OWL ontologies with axioms are relevant but not sufficient. - [11:20] TillMossakowski: sorry, just have lost my Skype connection when you asked your question, Christoph [11:21] ChristophLange: @TillMossakowski: my question was, as you certainly believe in the relevance of (formal) ontology languages, whether you can imagine applications powered by big data, which the integration of ontology languages enables [11:21] GaryBergCross: @ChristophLange +1 on your job of asking interesting, generative and relevant questions to our speakers. [11:25] AmandaVizedom: @OliverKutz, perhaps it would be appropriate to mention the summit OntoHub here, as somewhere Summit participants could go to try out these tools. [11:25] GaryBergCross: Isn't there a Protege plug-in for UML? Seems I heard of this. [11:25] TillMossakowski: I would be interested... [11:27] GaryBergCross: My [11:25] comment "The UML back-end plug-in provides an import and export mechanism between the Protege knowledge model and the object-oriented modeling language UML. see http://protegewiki.stanford.edu/wiki/UML_Backend [11:26] PeterMidford: @GaryBergCross - yes, but I think it is for Protege Frames, not the same as OWL. [11:26] MikeBennett: Re: Making OWL and UML inter-operable: is this because UML provides a better means of providing visualization to business domain folks, or is it because you actually need a model in the OO formalism of UML? If it's the former, why not use the ODM profile which lets you render OWL constructs in a UML tool? [11:27] TillMossakowski: it is because OWL is used to ontological modeling, and UML for software engineering [11:27] TillMossakowski: these are two related but different things (I mean ontology and software engineering) [11:31] MikeBennett: @Till thanks. [11:27] EricChan: Please include EricChan (I believe KenBaclawski will be interested) in the proposed offline discussions on translation between UML and OWL. Ken and I have worked on this for OASIS ICOM model. [11:38] TillMossakowski: @EricChan: works on OWL<->UML [11:39] MikeBennett: I would also like to be copied in on any UML to OWL mapping activities [11:28] JohnMcClure: my question for AlanRector is this: please discuss the impact on the vocabulary of properties, as a result of reifying properties as classes -- does it essentially remove nouns from the domain of properties, ending with just verbs + prepositions as properties??? [11:28] AmandaVizedom: @ChrisWelty: is Wikidata being used for Watson at all? [11:30] TillMossakowski: @ChrisWelty: but for typing, you use subclass hierarchies, if I remember correctly? [11:31] TillMossakowski: and then you could also use inferred subclass hierarchies? [11:31] GaryBergCross: (referring to MichaelGruninger's reference to "Spock" in his verbal comment about ChrisWelty's remark) Next IBM advance would be called "Spock"??? [11:32] MichaelGruninger: @GaryBergCross: no, my motto is "What would Spock do?" [11:32] PeterYim: Please mark you calendars and reserve this time, every Thursday, for the OntologySummit2014 virtual panel session series.In particular ... Session-04 will be up next Thursday- Thu 2014.02.06 (same time) - Track C: Overcoming Ontology Engineering Bottlenecks-I - see developing details at: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2014_02_06 [11:36] AnatolyLevenchuk: @all Remember about Hackathon event (track E): suggest your project ( http://ontolog.cim3.net/cgi-bin/wiki.pl?OntologySummit2014_Hackathon )! [11:32] EdBernot: Thanks all! [11:33] GaryBergCross: @ChristophLange, Thanks again for doing such a great job getting these 3 speakers and managing the session! [11:33] ChristophLange: @GaryBergCross: thank you! [11:33] ChristophLange: @PeterYim: Do we have a time limit? [11:33] PeterYim: feel free to go on for another 5 minutes (10 min. max) [11:37] SimonSpero: One place where KR and Big Data can be useful can be in aggregating and classifying instances, so that less data needs to be captured for each data point, and subgroups can be aggregated to get big enough counts for significance [11:40] SimonSpero: For example if one wanted to estimate life times for optical media in real situations, then some attribute may be constrained by manufacturer (e.g. dye types) [11:39] PeterYim: Please mark you calendars and reserve this time, every Thursday, for the OntologySummit2014 virtual panel session series.In particular ... Session-04 will be up next Thursday - Thu 2014.02.06 (same time) - Track C: Overcoming Ontology Engineering Bottlenecks-I - see developing details at: http://ontolog.cim3.net/cgi-bin/wiki.pl?ConferenceCall_2014_02_06 [11:40] GaryBergCross: Great session and lots to think about. The mail discussion should be useful. [11:40] PeterYim: -- session ended: 11:38 am PST -- ------