Ebook: Formal Ontology in Information Systems

This volume contains the papers presented at the 10th edition of the Formal Ontology in Information Systems conference, FOIS 2018, held between September 17–21, 2018, in Cape Town, South Africa. The conference was preceded by the 4th Interdisciplinary School on Applied Ontology, ISAO 2018, held between 10–15 September 2018 and was at the centre of a series of initiatives: the Joint Ontology Workshops 2018 (JOWO 2018) on September 17–18, the Early Career Symposium (ECS 2018) on September 19, and two tutorials, one on Conceptual Ontology Engineering and one on Referring Expressions in Ontologies and Query Answering.

The Joint Ontology Workshops 2018 comprised five specialised workshops, namely: Cognition And OntologieS 3 (CAOS-CEX), Epistemology in Ontologies II (EPINON II), the 6th International Workshop on Ontologies and Conceptual Modelling (Onto.CoM), Ontology of Economics, as well as the BadOntoloGy (BOG) workshop. Also, the Early Career Symposium was held within the FOIS conference program to give students and starting researchers the opportunity to present their work and engage with senior scientists in their field as well as to meet and discuss their work with each other in an open and friendly environment.

We received 48 submissions to the main track of the conference, 4 submissions to the ontology competition, and 6 submissions to the Early Career Symposium, coming from 20 countries. All papers were peer reviewed by at least three members of the programme committee (listed below). Based on the reviews we received, we accepted fourteen full papers (29,2% of submissions) and five short papers for the main track. Two of the ontology competition submissions were accepted for presentation during JOWO, and the Early Career Symposium accepted five submitted proposals for presentation as posters and lightning talks in a dedicated session of the conference.

As it usually happens at FOIS conferences, the variety of topics, methods, results, formalisms etc. one can find in the papers is astonishing. We grouped the contributions to the main track into four broad categories and organised them in respective chapters of the proceedings as follows:

• Foundations (7 papers)

• Agents and Properties (4 papers)

• Methods and Tools (4 papers)

• Applications (4 papers)

Regarding applications of ontologies, also a broad spectrum of areas is covered, including in particular biology and medicine, IoT, engineering and linguistics.

Winners of the FOIS Best Paper Award were announced during the conference. Awards and runners-up can be found at iaoa.org/fois/.

Acknowledgements

In this edition of the conference, we have received the support of several organisations. In particular, we would like to thank the University of Cape Town for supporting the organisation of the ISAO summer school with free rooms and infrastructure. Special thanks go to C. Maria Keet for her excellent work as a Local Chair for FOIS 2018, as well as to Zubeida Khan for her help with the ISAO 2018 website. We also thank IOS Press for sponsoring the FOIS Best Paper Award.

FOIS/JOWO/ISAO 2018 were organised under the auspices of our scientific promoter association, the International Association for Ontologies and its Applications (IAOA – iaoa.org).

Last but not least, we would like to thank our three invited speakers, Riichiro Mizoguchi, Alessandro Oltramari and Peter Simons for delivering keynotes at the conference.

Stefano Borgo

Pascal Hitzler

Oliver Kutz

Ontologists build formal models to understand the structure of reality. The fun starts –and I had a lot of it back in the PhD days (!)– when Formal Ontology is applied to understand the structure of what we indisputably use to understand reality itself: the mind. Philosophers have spent lifetimes hovering over this conundrum but I stopped more than a decade ago. Fast-forwarding to the present, I have been busy with a not-so-distant, yet more mundane, problem: building ontologies for AI.

My work focuses on engineering ontologies that can be integrated with the artificial minds' “substrata”, i.e. deep and shallow neural networks, and with the processes these bring about, all of which pretty much boil down to pattern recognition. In this keynote I will describe how ontologies can be effectively used in data-driven AI frameworks: I will argue that, in order to progress towards Explainable AI, it is necessary to design hybrid systems that integrate human-accessible machine representations with neural machines. Rather than concocting a philosophical theory, I will build my argument by illustrating core results from some of the projects I have been involved in at Carnegie Mellon first and, more recently, at Bosch.

While working on ontology engineering for about 25 years, I think I unconsciously acquired my own methodology in ontology engineering. It is what I call ‘state-centric’ methodology. My background is in AI and Engineering rather than logic or philosophy, and it must have been driving me to understand reality in terms of states which are the heart of General Problem Solver (GPS). All my results on roles, functions, processes/events and causation (published in Applied Ontology) clearly show the utility of the state-centric methodology. My treatment of context seems rather unique since I claim that any object as a whole provides a context to its parts, and each of these play their assigned role to make the collection a unitary whole. The notion of ‘non-intentional goal’ together with the systemic context as its concretization works very well to capture temporal entities. The success of the unified definition of biological and artifact functions is largely based on this idea. The device ontology, which I have devised for enabling to capture dynamic and complex phenomena in a consistent manner, has been exploited in many of my works on function. It is a role assignment system and fully state-centric. In my recent work on causation, I have realized that any causation C → E can be mapped onto a function in which E is a non-intentional goal in the systemic context associated with C → E. The new functional talk of causation is fully state-centric which contrasts with the common idea that states should be excluded from the relata of causation. I found that the essence of causation exists in an occurrent in the form <Event cause State> where State is the resultant state of the Event. In my keynote, I discuss the state-centric methodology through typical achievements on roles, function, processes/events and causation.

Perceptions, thoughts, pictures and expressions are all typically about something. In mental cases the relation is called intentionality, in pictures depiction, in expressions denotation. The nature of this aboutness has long been a topic for philosophical puzzlement and controversy. Whether it requires the existence of a kind of thing, quality or relation not found in inanimate nature, whether it is analysable, whether it comes in one or many forms, are all matters of dispute. The ontology of aboutness has to be at least plausibly conjectured if its features are to be represented within information systems that are sophisticated and capable enough to themselves represent representation. It falls therefore to the ontologist to investigate the entities and factors required and suitable to capture the form and matter of representation. This is no straightforward task, and there are many pitfalls. But it is a task that must be taken up if ontologies and the information systems that employ them are to advance to a stage where conjecture, diversity of opinion, (mis)information, uncertainty, falsehood, error, revision, contradiction and correction are to be smoothly represented and reconciled, and linked, as they must be, to action and decision, whether natural or artificial. Looking for help past and present, this paper sets about addressing that difficult task.

Particular types are designated in this paper as types dependent on an individual, such as SpouseOfHenryVIII dependent on HenryVIII. Other notable possibilities include car models, biological species, and various geological formations. A characterization and formal representation is provided for particular types that (1) introduces particular dependence between a type and an individual; (2) is grounded in this dependence and some defining relation for the type; and (3) provides a multi-level ontology pattern, using lakes as exemplars. This expands the range of types available to geographical ontology and beyond.

It is generally agreed that as a fundamental feature of the world, time merits treatment within an upper ontology, that is, an ontology that is designed to capture those categories which are sufficiently generic to transcend the specific subject matters of any particular domain ontologies. In this paper I examine how time is handled within three well-known upper ontologies (BFO, DOLCE, and GFO), and follow this with a discussion of three key issues emerging from the survey, namely dimensionality (the treatment of instants and intervals), frame-dependence (as required by the Theory of Relativity), and indexicality (the status of past, present and future). The overall conclusion is that while existing upper ontologies tend to adopt some kind of compromise between a supposedly objective, scientific account of physical time, and a more subjective, cognitive account of time as we experience it, the fundamental philosophical and scientific questions concerning the nature of time have scarcely been addressed by any of them.

The paper provides a tentative formulation of the diachronic identity criteria for localities based on a set of paradigmatic case studies of changes they may undergo.

A concept is traditionally defined via the necessary and sufficient conditions that clearly determine its extension. By contrast, cognitive views of concepts intend to account for empirical data that show that categorisation under a concept presents typicality effects and a certain degree of indeterminacy. We propose a formal language to compactly represent concepts by leveraging on weighted logical formulas. In this way, we can model the possible synergies among the qualities that are relevant for categorising an object under a concept. We show that our proposal can account for a number of views of concepts such as the prototype theory and the exemplar theory. Moreover, we show how the proposed model can overcome some limitations of cognitive views.

Structural universals have been introduced in the philosophical literature on examples such as chemical molecules composed of bonded atoms of different kinds. They are useful to handle complex abstract entities such as designs of artefacts, shapes, strings, words and texts, relevant in many areas of applied ontology. We use graph-theory as a unifying framework to review and compare the formal accounts proposed by Armstrong, Bennett and Mormann. We then propose a more expressive account, combining features of Bennett's and Mormann's proposals, able to model both the mereology of complex states of affairs and the structure of the particulars involved in them. Structural universals are explicitly represented; their structure, where a same universal can be part of a structural universal several times over, can be read off the graphs representing complex states of affairs.

The idea that the real-world entities referred to by Information Systems are determinate and uniquely identifiable is a commonly held assumption in the fields of Software Engineering and Database Systems. The concept of identity is also a central topic in Formal Ontology, a discipline that finds application in the field of Information Systems through the use of Foundational Ontologies (FOs). However, while most central concepts of Formal Ontology are, in general explicitly addressed in FOs, the concept of identity has received relatively little attention. The lack of a proper ontological characterization of identity in FOs hinders their application to the analysis of issues related to identification in Information Systems, such as those that arise in conceptual modeling or in database design. This work proposes two distinct, but logically equivalent, formal characterizations of the notion of individual determinacy. Moreover, these characterizations are independent of the particularities of a FO's theory and are defined solely in terms of the structure of the FO's models of portions of reality. Finally, it also introduces a few concepts that are useful in the analysis of identity criteria for the individuals represented using a FO's theory.

It is widely recognized that accurately identifying and classifying competitors is a challenge for many companies and entrepreneurs. Nonetheless, it is a paramount activity which provide valuable insights that affect a wide range of strategic decisions. One of the main challenges in competitor identification lies in the complex nature of the competitive relationships that arise in business environments. These have been extensively investigate over the years, which lead to a plethora of competition theories and frameworks. Still, the concept of competition remains conceptually complex, as none of these approaches properly formalized their assumptions. In this paper, we address this issue by means of an ontological analysis on the notion of competition in general, and of business competition, in particular, leveraging theories from various fields, including Marketing, Strategic Management, Ecology, Psychology and Cognitive Sciences. Our analysis, the first of its kind in the literature, is grounded on the Unified Foundational Ontology (UFO) and allows us to formally characterize why competition arises, as well as to distinguish between three types of business competitive relationships, namely market-level, firm-level and potential competition.

Clear criteria for the identity of dispositions are still lacking, and this has been presented as one of the main challenge raised by such entities. It is of prime importance to identify or distinguish dispositions such as diseases or risks. This article first introduces conventional ways to refer to a disposition (such as “fragility”) and canonical ways (such as “disposition to break in case of a strong shock”). This raises the issue of how should exactly be defined a “disposition d to R when TR”, where R is a realization specification and TR a trigger specification. Two ontological frameworks are distinguished. The first framework, which has been largely used so far in the literature on dispositions, interprets d as a disposition which can only be triggered by instances of TR, and can only be realized by instances of R. The second, new framework introduces the notion of “minimal trigger” and “maximal realization”, and interprets TR as a parent class of a class of processes that have as part a minimal trigger, and R as a parent class of a class of processes that are parts of a maximal realization. We then discuss several criteria of identity, including the criterion according to which two dispositions are identical iff they have the same categorical basis, the same class of minimal triggers and the same class of maximal realizations. We show on several examples that the second framework avoids the disposition multiplicativism that is introduced by the first framework.

How can organisations survive not only the substitution of members, but also other dramatic changes, like that of the norms regulating their activities, the goals they plan to achieve, or the system of roles that compose them? This paper is as first step towards a well-founded ontological analysis of the persistence of organisations through changes. Our analysis leverages Kit Fine's notions of rigid and variable embodiment and proposes to view the (history of the) decisions made by the members of the organisation as the criterion to re-identify the organisation through change.

We discuss steps towards a formalisation of the principles of an agentive naïve proto-physics, designed to match a level of abstraction that reflects the pre-linguistic conceptualisations and elementary notions of agency, as they develop during early human cognitive development. To this end, we present an agentive extension of the multi-dimensional image schema logic ISL based on variants of STIT theory, thus replacing the temporal dimension of ISL with an action-agnostic theory of agency. To begin grasping the notion of ‘animate agent’, we apply the newly defined logic to model the image schematic notion of ‘self movement’ as a means to distinguish the agentive capabilities of a mouse from those of a ball. Finally, we outline the prospects for employing the theory in cognitive robotics.

Engineers distinguish between nominal and actual qualities, a dichotomy that is fundamental to guarantee that physical objects satisfy design requirements. Computational ontologies are broadly exploited across engineering domains, even though they do not attempt at making explicit the intended semantic of the two notions. The purpose of the paper is to present a foundational analysis of nominal and actual qualities to support their robust specification for knowledge representation. Instead of presenting an ontology, we discuss various modeling alternatives on which users can rely to develop their ontologies.

This paper presents Amnestic Forgery, an ontology for metaphor semantics, based on MetaNet, which is inspired by the theory of Conceptual Metaphor. Amnestic Forgery reuses and extends the Framester schema, as an ideal ontology design framework to deal with both semiotic and referential aspects of frame and role mappings. The description of the resource is supplied by a discussion of its applications, with examples taken from metaphor generation, and the referential problems of metaphoric mappings. Both schema and data are available from the Framester SPARQL endpoint.

Mereotopological relations, such as contact, parthood and overlap, are central for representing spatial information qualitatively. While most existing mereotopological theories restrict models to entities of equal dimension (e.g., all are 2D regions), multidimensional mereotopologies are more flexible by allowing entities of different dimensions to co-exist. In many respects, they generalize traditional spatial data models based on geometric entities (points, simple lines, polylines, cells, polygon, and polyhedra) and algebraic topology that power much of the existing spatial information systems (e.g., GIS, CAD, and CAM). Geometric representations can typically be decomposed into atomic entities using set intersection and complementation operations, with non-atomic entities represented as sets of atomic ones. This paper accomplishes this for CODI, a first-order logic ontology of multidimensional mereotopology, by extending its axiomatization with the mereological closure operations intersection and difference that apply to pairs of regions regardless of their dimensions. We further prove that the extended theory satisfies important mereological principles and preserves many of the mathematical properties of set intersection and set difference.

This decomposition addresses implementation concerns about the ontology CODI by offering a simple mechanism for determining the mereotopological relations between complex spatial entities, similar to the operations used in algebraic topological structures. It further underlines that CODI accommodates both quantitative/geometric and qualitative spatial knowledge.

The W3C standardized Semantic Web languages enable users to capture data without a schema in a manner which is intuitive to them. The challenge is that, for the data to be useful, it should be possible to query the data and to query it efficiently, which necessitates a schema. Understanding the structure of data is thus important to both users and storage implementers: The structure of the data gives insight to users in how to query the data while storage implementers can use the structure to optimize queries. In this paper we propose that data mining routines be used to infer candidate n-ary relations with related uniqueness- and null-free constraints, which can be used to construct an informative Armstrong RDF dataset. The benefit of an informative Armstrong RDF dataset is that it provides example data based on the original data which is a fraction of the size of the original data, while capturing the constraints of the original data faithfully. A case study on a DBPedia person dataset showed that the associated informative Armstrong RDF dataset contained 0.00003% of the statements of the original DBPedia dataset.

Research and Innovation (R&I) ecosystems involve data and knowledge flows across enterprises, academia, funding institutions, public authorities and citizens. Key R&I data elements are currently dispersed across a multitude of distinct and heterogeneous datasets; they are often neither in structured format nor systematically shared. Designing policies that affect R&I dynamics, and monitoring their impact in time, are tasks which require to overcome the limitations imposed by the usage of separated data silos, and to provide meaningful, integrated access to data with the appropriate granularity. In such a context, ontology-mediated data management can help bringing together inputs and outcomes from a variety of sources, in an open and interoperable fashion. The paper introduces UNiCS, an OWL-based ontology that has been recently developed with the main aim of driving semantic integration and access to R&I data, and to support the development of ontology-based data access (OBDA) platforms for R&I policy and decision makers.