Basic Formal Ontology Summit Meeting/abstracts

From NCOR Wiki
Jump to navigationJump to search

Day 1: Future of Basic Formal Ontology

  • Werner Ceusters: The Axiomatization of BFO 2020: The Good, The Bad, and The Ugly

The Basic Formal Ontology (BFO) is undoubtedly the most frequently cited upper ontology in the biomedical literature. It is also claimed to be used as the foundation for many biomedical application ontologies. However, a closer look at these ontologies makes it clear that many of them deviate considerably from even the most basic principles upon which the BFO is built. Reasons can be derived from published papers that describe these ontologies. One reason is the popularity of OWL and tools such as Protégé of which the limits are poorly understood. A second one is the loss of semantics that arises from representing the BFO exclusively by means of binary relations, thus ignoring certain principles underlying the BFO, notably the requirement for temporal indexing in relations involving continuants. Although the importance of these and other principles has been stressed in many papers and presentations by BFO authors, education therein seems to be lacking or is at least inadequate.

The recent axiomatization of the BFO in First Order Logic (FOL) might offer an opportunity to remediate the situation. Soon after its public availability in September 2021, I started to use the BFO’s CLIF files in my ontology and referent tracking courses to explain BFO’s foundational principles. For some assignments, students were expected to write axioms for terms and relations they intended to include in the ontologies they were developing as part of their MSc or PhD thesis. I also embarked on a still ongoing project to develop a set of tools to make axiom-based ontology design easier for students lacking a solid background in formal logic and/or computer science. It is therefore encouraging to see that others are following similar paths and, for example, have shown that automatic reasoning with FOL annotations can be used to detect previously unnoticed errors in simple OWL-based classifications.

In this talk, I will focus on some of my and my students’ experiences with the use and development of a parser/generator for BFO2020-style CLIF axioms, and a reasoner for satisfiability testing of both axiomatized ontologies and instance data expressed in their terms. I will discuss how these experiences – good ones (yeah, temporal reasoning becomes possible!), bad ones (is there a bug in the reasoner or are temporal regions not what I thought them to be?), and ugly ones (is the resulting model one that we intended?) – might be used to improve the documentation of the BFO, to make axioms more actionable and informative while still logically equivalent and to provide a smoother path to axiomatization of related ontologies such as the Ontology of General Medical Science, the Information Artifact Ontology and the Ontology of Biomedical Investigations.

  • Fabian Neuhaus: BFO-FOL for Domain Ontologies

The term "Basic Formal Ontology" is ambiguous. Firstly, it may denote some ontological position in an Aristotelian tradition combined with some methodological principles for the development of domain ontologies (BFO-Phil). Secondly, it may refer to a document, which represents the philosophical theory in first-order logic (or Common Logic) axioms (BFO-FOL). Thirdly, it may refer to a document, which represents the philosophical theory in OWL DL (BFO-OWL). (Both BFO-FOL and BFO-OWL exists in several variants.) BFO-Phil, BFO-FOL, and BFO-OWL differ dramatically with respect to their content. BFO-Phil builds on a rich philosophical tradition and has been developed over many years in numerous articles, a user guide and a text book. The various variants of BFO-FOL capture only a small, but a significant fraction of the content of BFO-Phil. In contrast, BFO-OWL is so weak that almost none of BFO-Phil is preserved. For example, BFO 2.0 OWL, which is used by many domain ontologies, contains exactly 52 OWL axioms, more precisely 18 disjointness axioms and 34 subsumption axioms between atomic classes. Consequently, the rich ontological content of BFO-Phil is not used for automatic reasoning and consistency checking. FOWL is a tool that was developed to support reasoning with ontologies that consists of both OWL and FOL axioms. One particular use case of FOWL is the integration of BFO-FOL and BFO-OWL, because it enables domain ontologists to develop their ontologies in OWL (and use OWL reasoners for their purposes) and in addition to use FOL-axioms of BFO for validation purposes.

  • "Time and the Continuant. Modeling Temporal Data in BFO-OWL", Ludger Jansen

In its latest incarnation, BFO opted for ontological relations with various temporal strengths (at all times vs. at some time). This was an enormous step forward as compared to the temporally unenlightened relations used before. Since some time, I am considering an alternative modelling approach for BFO-OWL that tries to integrate temporal information into the continuant class. The talk introduces and defends the underlying idea, presents the technical apparatus, and suggest some possible applications.

Day 2: Applications of Basic Formal Ontology

  • Janna Hastings, Field Reports from Using BFO in Scientific Ontologies
  • "The Industrial Ontologies Foundry", TBD
  • "Fandaws and BFO", Aaron Damiano
  • "OccO: The Occupation Ontology" John Beverley, Yongqun "Oliver" He, Sam Smith, et al.

Though there is widespread agreement on the importance of representing occupational data, taxonomy standards that exist - e.g. ILO ISCO, EU ESCO, US BLS SOC - are not interoperable, despite covering roughly the same domain. Ontologies - controlled vocabularies comprised of classes arranged in a hierarchy and formal relationships among them - have for many years been deployed to promote interoperability among disparate datasets. With this in mind, and with the three standards referenced above in purview, we have developed the Occupation Ontology (OccO), whose scope is specific to the domain of occupations, titles, relevant skills and capacities. Though currently focused on English language occupation standards, next steps include extending coverage of OccO to ISCO and ESCO, and indeed our progress has been enhanced by dialogue with representatives of the latter. As expansions of coverage continue, moreover, the OccO development team will maintain conformance to the ISO specified upper-level Basic Formal Ontology, as well as to the principles of the Open Biological and Biomedical Ontology Foundry, as such conformance results in interoperability among a wide range of existing conformant ontologies.

  • "BFO Applications in Interoperable Biomedical Ontology Development and Knowledge/Data Integration – He Lab Experience", Yongqun "Oliver" He

The Basic Formal Ontology (BFO) is an upper level ontology that has been widely used in interoperable biomedical ontology development and supports biomedical data and knowledge integration and analysis. This talk presents our own He lab experience in usage and applications of the BFO for our biomedical research. Our initial application of BFO came from our development of the community-based Vaccine Ontology (VO), which reuses over 20 BFO-aligned ontologies (e.g., OBI, GO, and IDO) and develops vaccine-specific models using the BFO framework. Later we have initiated many other reference ontologies such as the Ontology of Adverse Events (OAE), Cell Line Ontology (CLO), Ontology of Precision Medicine and Investigation (OPMI), and Coronavirus Infectious Disease Ontology (CIDO). We have also developed application ontology: Kidney Tissue Atlas Ontology (KTAO). All these ontologies are aligned with the BFO, supporting seamless ontology interoperability. BFO-based ontology supports knowledge integration, sharing, query, and analysis. For example, VO and CIDO are used to represent and integrate various knowledge related to vaccine and COVID-19, respectively. Furthermore, BFO- based ontologies also support data integration. For example, OPMI and KTAO have also been developed to support the NIH-funded Kidney Precision Medicine Project (kpmp.org), which involves over 20 universities in the US in a collaborative effort to generate and process a large volume of heterogeneous clinical, molecular, and pathological data. We are also using the BFO ontology framework to model and represent uncertainty and its related entities, and further apply the model to study the question of COVID-19 vaccine hesitancy. Also noted that we have developed ontology-based tools to support BFO- aligned ontology development and applications.

  • "Ontologies for Sustainability: Theoretical Challenges", Giorgio Ubbiali, Nicolas Piras, Matthew Lange, Andrea Borghini

To date, no study has reviewed the available ontologies with regard to the notion of sustainability and the connected theoretical challenges. This paper aims to fill this gap. First, in section 1, we outline the three major challenges associated with the notion of sustainability: 1) the polysemy of the term, 2) the relationship between sustainability and sustainable development, and 3) the complexity of the topics covered by this notion. Then, in section 2, we offer a review of the main accomplishments, achieved so far by ontologies, to meet these theoretical challenges. In this assessment, we devote special attention to the Sustainable Development Goals Interface Ontology (SDGIO). SDGIO is the only available ontology that focuses on sustainability and employs the Basic Formal Ontology (BFO) as the upper-level ontology. Since, within the OBO Foundry, all ontologies utilize BFO, this facilitates sharing and integrating terms systematically, allowing the establishment of richer accounts of notions as challenging as sustainability. We conclude by considering some potential trajectories to improve SDGIO and further develop a family of sustainability ontologies, which employ BFO as the reference upper-level ontology.

  • "Ontologies for a Learning Health System: Application to PARS3 and theoretical challenges", Adrien Barton, Paul Fabry, Jean-François Ethier

The development of a Learning Health System in Quebec named "PARS3" raises the issue of interoperability between various clinical data sources. To address those, we have developed a series of ontologies of informational entities and processes in the clinical domain based on the OBO Foundry methodology. We will give an overview of those core ontologies of clinical informational entities, including the Prescription of Drugs Ontology (PDRO), the Laboratory Ontology (LABO), the Demographic data Ontology (DEMO) and the Questionnaire Ontology (QUESTO). We will show how they have been recently updated and integrated into a wider frame including healthcare systems (HADO), health procedures and health activities (HEPRO).

We will also expose two theoretical issues raised by such ontologies. First, how directive informational entities (e.g. the prescription of a laboratory test) can be related to the processes they may direct (e.g. a laboratory test). Second, how informational entities (e.g. a clinical document) can be described mereologically by a variant of slot mereology, according to which the same informational entity can have several times the same part (e.g. the term ‘diabetes’ appearing several times on the same document).

  • "Accounting Contracts in Collaboration Space", Bill McCarthy and Graham Gal

This paper proposes a design theory for accounting transaction systems that is based on the resource-event-agent (REA) accounting ontology and distributed ledger technology (DLT). Following a design science research approach, we present a novel blockchain-based prototype that implements the REA ontology as a business process state machine. Our prototype utilizes the unique features of DLT to ensure data integrity and transparency among the economic agents in recorded and distributed repositories that reside outside of the enterprise systems of those parties. This provides evidence that the independent view of business transactions (as defined in ISO 15944-4 (2007) between and among companies and government agencies - in what we call collaboration space - is technologically feasible and opens avenues for exploring innovative ways for performing economic exchanges. Because our business transaction choreography is orchestrated by accountability contracts for reciprocated delivery of economic resources, we title our treatment “accounting contracts in collaboration space.”

  • "Towards an Ontology of Inflation", Jonathan Vajda

What is price inflation in an economy? I argue that price inflation is a change of disposition in an aggregate of personal, social agents. Currency inflation is itself one such example of price change for a given medium of exchange. My proposal might seem initially counter-intuitive, since prices are usually thought to be traits of things and services rather than agents. First, I give a service-centric account of commercial exchanges. Second, I give a dispositional account of price. Third, I argue that aggregates are the proper ontological elements of price inflation and currency inflation, and that some of the relevant processes (such as joint actions) are realizations of dispositions inhering in aggregates. The intuition that inflation is a property of things and services is, thus, reframed relative to an aggregate of persons. This account purports to advance an inchoate ontology, and does not suggest a complete etiology nor an appraisal of monetary policies that intend to affect inflation. I address several objections, including: that personal social agents (nor aggregates thereof) are neither necessary nor sufficient for economic valuation activities; that inclusion in an aggregate is arbitrary; that attributions of price are properly said of the thing or service; and other objections.

  • "Advantages and Challenges of Building a Realism-based Ontology of Finance", Gloria Sansò

Over the last decade, it has become evident that an ontology of finance is needed. Such an ontology would assist in many activities such as assessing systemic risk, detecting financial frauds, and ranking companies based on their financial quality. So far, two major attempts to provide this ontology have been made. The first has led to the Financial Industry Business Ontology (FIBO), whereas the second has led to the Ontology Network in Finance and Economics (OntoFINE). Both these ontologies are concept-based, meaning that they aim to represent concepts of reality instead of the reality itself. Building an ontology of finance BFO- conformant (and thus realism-based) would been an interesting alternative, especially considering that, on several occasions, BFO and its extensions have been proven to be more successful than some concept-based ontologies. On the other hand, financial entities are typically non-physical and, as such, they are a challenge for a framework designed to represent physical things. This point will be discussed by considering some examples including “bond”, “stock”, “price”, and “volatility”.

Day 3: Government Ontology

  • Keynote: Ryan Riccucci: For a Government Ontology
  • Buffalo and Toronto Ontology Group (BoaT) (Chair - Mark Fox)