Actions

Ontolog Forum

Session Topic Discussion
Duration 1 hour
Date/Time July 17 2019 16:00 GMT
9:00am PDT/12:00pm EDT
5:00pm BST/6:00pm CEST
Convener Ken Baclawski

Ontology Summit 2020 Topic Discussion

Agenda

The following agenda items may require multiple meetings.

  1. Discussion of the three topic areas
    1. Knowledge graphs Gary Berg-Cross
    2. Ontologies and standards Mike Bennett
    3. Analogy and Metaphor David Whitten (to be confirmed)
  2. Selection of the topic area
  3. Choose tracks and track champions
  4. Draft summit mission statement
  5. Choose tentative summit title

Ontology Summit Topic Suggestions

  • Knowledge graphs and ontologies
    • Example ref: Summarizing ontologies using graphs: A Survey [1] also [2] (both links are the same article)
    • Knowledge graphs have recently become popular
    • Unfortunately, much of this activity is proprietary with no real standards
    • The Google Knowledge Graph has been criticized for the lack of provenance
    • Champion (during planning sessions): Gary Berg-Cross
  • Ontologies and standards (revisiting Ontology Summit 2009)
    • Rationales for ontology based standards
    • What are ontologies for standards?
    • Is the focus ontologies for standards or standards for ontologies or both?
    • What is the relationship between NL and standards?
    • Ontological Standards: Past, Present, and Future is my preferred topic (Ram D. Sriram)
    • Start by talking about the existing approches e.g. use of terminologies, disctionaries and so on, and show where that leads (show some leadership). Lead in to broader topic of geetting formal common meaning (ontology) into industry standards, formalizing the knowledge that is already there. (Mike Bennett)
    • Presentation on Ontology Standards by Ravi Sharma
    • Champion (during planning sessions): Mike Bennett
  • Analogy and Metaphor
    • "Metaphor is for most people device of the poetic imagination and the rhetorical flourish--a matter of extraordinary rather than ordinary language. Moreover, metaphor is typically viewed as characteristic of language alone, a matter of words rather than thought or action. For this reason, most people think they can get along perfectly well without metaphor. We have found,on the contrary, that metaphor is pervasive in everyday life, not just in language but in thought and action. Our ordinary conceptual system, in terms of which we both think and act, is fundamentally metaphorical in nature." from "Metaphors We Live By" by George Lakoff and Mark Johnson 1980
    • "One of the main contentions of this book is that the cognitive mechanisms involved in metaphor may prove to be a better paradigm for structuring knowledge that that of first-order logic." from the Preface to "Knowledge Representation and Metaphor" by Eileen Cornell Way 1991
    • "Analogy, metaphor, representation and user interface have much in common: each involves signs, meaning, one or more people, and some context ..." "An introduction to algebraic semiotics, with application to user interface design" by Joseph Goguen 1999
    • Champion (during planning sessions): David Whitten

Conference Call Information

  • Date: Wednesday, 17-July-2019
  • Start Time: 9:00am PDT / 12:00pm EDT / 6:00pm CEST / 5:00pm BST / 1600 UTC
  • Expected Call Duration: 1 hour
  • The Video Conference URL is https://zoom.us/j/689971575
    • iPhone one-tap :
      • US: +16699006833,,689971575# or +16465588665,,689971575#
    • Telephone:
      • Dial(for higher quality, dial a number based on your current location): US: +1 669 900 6833 or +1 646 558 8665
      • Meeting ID: 689 971 575
      • International numbers available: https://zoom.us/u/Iuuiouo
  • Chat Room

Participants

Proceedings

[12:05] Gary: The compromise and trade-off values of KGs makes them attractive to some people. Their value proposition is partly based on this and reflected in this - "Knowledge graphs and similar structures usually provide a shared substrate of knowledge within an organization, allowing different products and applications to use similar vocabulary and to reuse definitions and descriptions that others create. Furthermore, they usually provide a compact formal representation that developers can use to infer new facts and build up the knowledge for example, using the graph connecting movies and actors to find out which actors frequently appear in movies together."

[12:06] David Eddy: Must read book: Murphy & Yates "Engineering Rules" history of standards setting efforts beginning about 1900.

[12:06] David Eddy: It's a looooooong slog.

[12:06] janet singer: Sounds like fun?

[12:10] David Eddy: What about the SYSTEMS that are NOT "out on the web?" The operational systems BEHIND the organizational firewall?

[12:11] Gary: Freebase: A Collaboratively Created Graph Database For Structuring Human Knowledge Kurt Bollacker, Colin Evans, Praveen Paritosh, Tim Sturge, Jamie Taylor

[12:11] David Eddy: I would argue the SemWeb "stack" totally ignores the decades of accumulated systems behind the firewall. TOTALLY inaccessible to Google slurping.

[12:11] TerryLongstreth: I'd like to see standards as sub-topic. I've long been associated with standards development, and am currently in a group performing conformance tests ("audits") of organizations w/r/t the ISO 16363 standard, which I helped write. But standards is an unlimited topic which covers, conceptually, all areas of human activity. So, necking down to standards appropriate to some other topic is not easy.

[12:13] David Eddy: To the best of my followings... John Sowa is the only person here who raises the issue of dealing with operational systems.

[12:16] David Eddy: Huge challenge (from CSAIL observations)... academics have no access to production systems.

[12:21] Gary: We shouldn't do anything on standards without first considering what was done in OntologySummit2009: Toward Ontology-based Standards [3]

This is a joint initiative by OASIS, OMG, various ISO working groups, IAOA, OOR and ONTOLOG that is spun off from OntologySummit2009.

This joint initiative is a collaborative effort by the co-organizers to drive the Ontology-based Standards agenda among the Standards Development community, the Ontology community and the Systems community. Moving "Toward Ontology-based Standards" was adopted as the theme for the Ontology Summit in 2009. Some of the projects started back then, plus others that have emerged since, have produced encouraging results. We believe that sharing the experiences gained from some of those projects with the aforementioned communities will go a long way towards helping move the Ontology-based Standards agenda forward and grounding the arguments for the adoption of the "Ontology-based Standards" approach.

With the above in mind, we are producing a mini-series of virtual panel sessions. The first two session will be showcasing efforts from the OASIS ICOM TC (on 25-Oct-2012) and those from OMG (on 8-Nov-2012.) In these two sessions, we are planning to showcase exemplary work where standards have been represented both in a natural language (English, in our cases) as well as in formal ontology language(s). Through the sharing of their experiences, their approaches and processes, we shall discuss potential benefits, issues, challenges and other learnings from those projects.

... see also, the OntologySummit2009_Communique.

[12:21] TerryLongstreth: Examples: Standards governing the development, capture, quality control, life-cycle and retirement; use in inter-human cooperative actions, use in automated interoperability environments

[12:25] David Eddy: Realistically... 10 years is not a very long time in the scheme of the complexity in this "terminology organization" domain. As a simple punch card programmer, I've been tilting at this since 1980.

[12:25] David Eddy: ISO 11179? Ever used in practice?

[12:25] RaviSharma: Gary reviewed units, measures, products, geospatial, semantics were topics, etc featured in 2009 summit

[12:28] RaviSharma: review of 2009 and now standards

[12:29] RaviSharma: showing path of progress in standards, standards that use ontologies then and now. etc.

[12:31] RaviSharma: KGs have continuity in last 3 summits as these topics were discussed

[12:32] David Eddy: Google MT approach ignores "un-natural language" embedded in operational systems

[12:32] David Eddy: "statistical approach"

[12:32] RaviSharma: OKN and Knowledge graphs have continuity in last 3 summits - Gary

[12:33] TerryLongstreth: Can we explore how Knowledge graphs support or impede use of Analogy and Metaphor?

[12:34] TerryLongstreth: RDBMS are not simply tables; constraints and triggers are essential parts of full implementations.

[12:36] TerryLongstreth: Constraints and triggers can be stated in statistical/fuzzy (but not ambiguous) forms

[12:36] RaviSharma: David Eddy - Google uses Statistics.

[12:36] Mark Underwood: In this (I think blog) post, there is at least a reference to ontologies... What bolsters the argument for KG's as a topic is the healthy industry activity in the space, where we'd like to at least have a modest role

[12:36] Mark Underwood: [4]

[12:37] TerryLongstreth: We may need to explore representation or standardization of "maybe" ontological commitments.

[12:38] RaviSharma: David- Machine translations use non-natural Languages.

[12:39] David Eddy: Unless you have the Captain Decoder ring... I defy you for MT to grok the meaning of M0101

[12:39] David Eddy: Where does the content in Knowledge Graphs come from?

[12:39] RaviSharma: Donna - agree with Gary on KG as a topic

[12:40] TerryLongstreth: reddis - Donna recommends

[12:40] RaviSharma: Donna - Reddis

[12:43] David Eddy: It's a pun.... same as "Engineering Rules"

[12:43] ToddSchneider: [5]

[12:43] ToddSchneider: Shapes Constraint Language (SHACL)

[12:45] David Eddy: Do remember... Tim Berners-Lee's intent was "linked data" (bad marketing)... so the label became "Semantic Web" Way sexier.

[12:46] ToddSchneider: Graphs, Knowledge Graphs, Ontologies

[12:46] Mark Underwood: Paper is behind a paywall for me, but ...

Complex Network Based Knowledge Graph Ontology Structure Analysis

Ontology is the core of knowledge graph. Traditional ontology description and ontology representation rely on ontology descriptional language. This kind of representation method makes it difficult for people to quickly grasp ontology's structure and then reuse it or segment it. To solve this problem, we proposed a method to transform ontologies into complex networks. This paper analyses ontologies' structural characteristics through ontology visualization and ontologies' degree distribution, clustering coefficient, average path length and eigenvector centrality. We observed that many ontologies have tree-like structures. Our analyses further revealed that a concept's importance is positively related to its degree and eigenvector centrality. Experiments in university ontology shows that our method has a good effect in intuitively understanding the ontology structure.

[12:46] Gary: [7]

[12:50] RaviSharma: Ken - John Sowa has ... on standards?

[12:50] Gary: There are many, many musing on the relation of KG and Ontologies. As I said you can use an ontology or several to structure the KG. Also there is the element of size. " Many would agree that sheer scale is part of what sets an ontology apart from a knowledge graph. Ontologies are generally regarded as smaller collections of assertions that are hand-curated, usually for solving a domain-specific problem. By comparison, knowledge graphs can include literally billions of assertions, just as often domain-specific as they are cross-domain."

[12:50] Ken Baclawski: @Ravi: [8]

[12:51] David Eddy: >> assertions that are hand-curated << That will never fly.

[12:51] Gary: More from the literature - "Many would argue that the divide between ontology and knowledge graph has nothing to do with size or semantics, but rather the very nature of the data. For example, dividing all class structures and relationship definitions into one group and all instance-level data into another might fulfill their idea of an ontology and knowledge graph, respectively one to be used for inference, and the other to be queried for examples. Its unlikely that a consensus will emerge anytime soon on what a knowledge graph is or how it is different from an ontology. For now, its more helpful to remember that the two approaches to are fundamentally the same. Most caveats stem from disagreements about size, the role of semantics and the separation of classes from instance data. But when it boils right down to it, they are generally larger or smaller versions of each other, with more or less sophisticated knowledge encoding techniques under the hood."

[12:52] David Eddy: But, but... finance is a fashion driven industry.

[12:52] RaviSharma: Neo4J was mentioned

[12:53] RaviSharma: Gary - Schema.org too general

[12:53] Mark Underwood: Possible theme title: "Shootout at the AI Corral: Ontologies vs. Knowledge Graphs"

[12:54] Mark Underwood: Gary suggests postponing the title, or treating as a working title

[12:54] David Eddy: After flogging his "Framework" for 25+ years (co author in IBM Systems Journal, John Sowa), John Zachman declared his framework an "ontology."

[12:54] janet singer: @David - I just bought the Engineering Rules book

[12:55] RaviSharma: Gary - I agree with you that we need to come up on title by studying a subject as a mini series, can we do both Knowledge Graph and role of standards and ontologies?

[12:55] David Eddy: @Janet... authors are spouses... he @Wellesley, she @MIT

[12:56] Mark Underwood: Or, pun-friendly "Ontologies or Knowledge graphs: What to draw upon"

[12:58] Mark Underwood: There's value in a modestly curated collection of use cases, which I found valuable in this past symposium

[12:59] Gary: A useful background article on KGs and Ontologies is this:

Ontologies for Knowledge Graphs? Markus Krötzsch

"Modern knowledge representation (KR), and description logics (DL) in particular, promises many advantages for information management, based on an unambiguous, implementation-independent semantics for which a range of reasoning services is available. These strengths align well with the needs of an ever growing information industry. Today, giants like Google, Facebook, and Wikimedia consciously deploy ontological models, and store information in graph-like data structures that are more similar to DL ABoxes than to traditional relational databases. Many smaller organisations follow, and knowledge graphs appear in numerous places. Clearly, logic-based KR can make significant contributions to this development, yet there is often very little adoption in typical knowledge graph applications. Focusing on Wikidata as a particular use case, this invited contribution asks which technical issues might limit the impact of symbolic KR in this area, and summarises some recent developments towards addressing them in various logics."

[9]

[13:02] Mark Underwood: Gary suggests a prequel meeting or two to refine the topic

[13:02] ToddSchneider: We shouldn't focus on only 'knowledge graphs', but graphs in general (i.e., graph databases).

[13:02] David Eddy: I see these discussions thusly (top to bottom) ontologies, taxonomies, naming standards. Top down & bottom-up MUST work together. Alone they're useless.

[13:03] David Eddy: See: (a late 1970s Zachman Framework implementation) http://www.tdan.com/view-articles/6123 See significance (entirely absent from W3C / SemWeb discussions) of footnote #11

[13:04] RaviSharma: Mark - developer world and high volume, APIs SPARC? open source etc. big data high volume etc

[13:05] Mark Underwood: [10]

[13:06] Mark Underwood: [11]

[13:06] RaviSharma: Reaching out to those in development world and exposing them to ontology is valuable

[13:06] Mark Underwood: [12]

[13:07] RaviSharma: Ken - plan a mini series on KG and Ontologies

[13:07] TerryLongstreth: @David: re:top vs bottom[ & how they are applied or can be governed to work together.]

[13:07] RaviSharma: Graphs and KG - Todd?

[13:07] David Eddy: @TS... "work together?" That will never fly. </sarc>

[13:09] TerryLongstreth: @Gary - re: more interest generated: who are we trying to attract?

[13:09] Mark Underwood: [13]

[13:09] RaviSharma: Todd - graph databases, KG and Ontologies, please expand

[13:10] RaviSharma: Ken suggests a mini series planning session.

[13:10] TerryLongstreth: Got to sign off - Will try to reconnect next week.

[13:11] ToddSchneider: Meeting ends @13:11 Eastern.

[13:13] RaviSharma: [14]

[13:14] RaviSharma: [15]

[13:23] Mark Underwood: @Ravi - Not sure we should give up on the visualization aspect you brought up. Almost every ontology introduction I've seen or given ends up with some sort of visualization. And it's not coincidentally tied to syntax trees, parse trees, Petri nets, decision tree.. It may be a sidebar topic, but perhaps important? E.g., this paper deals with visual perception, but touches upon parse trees as a usual concept.

[13:23] Mark Underwood: [16]

Resources

Previous Meetings

Next Meetings

... further results