Barry Smith

As engineering applications require management of ever larger volumes of data, ontologies offer the potential to capture, manage, and augment data with the capability for automated reasoning and semantic querying. Unfortunately, considerable barriers hinder wider deployment of ontologies in engineering. Key among these is lack of a shared top-level ontology to unify and organise disparate aspects of the field and coordinate co-development of orthogonal ontologies. As a result, many engineering ontologies are limited to their scope, and functionally difficult to extend or interoperate with other engineering ontologies. This paper demonstrates how the use of a top-level ontology, specifically the Basic Formal Ontology (BFO), greatly facilitates interoperability of multiple engineering-related ontologies. We constructed a system of formal linked ontologies by re-engineering legacy ontologies to be conformant with BFO and developing new BFO-conformant ontologies to capture knowledge in the engineering design, enterprise, human factors, manufacturing, and application domain of additive manufacturing. The resulting Integrated Framework for Additively Manufactured Products (IFAMP), including the body knowledge instantiated on its basis, serve as the basis for a proposed Design with Additive Manufacturing Method (DWAM), which we believe can support the design of innovative products with semantically enhanced ideation tools and enhanced access to application domain knowledge. The method and its facilitation through the ontological framework are demonstrated using a case study in medicine.

The goal of creating Artificial General Intelligence (AGI) – or in other words of creating Turing machines (modern computers) that can behave in a way that mimics human intelligence – has occupied AI researchers ever since the idea of AI was first proposed. One common theme in these discussions is the thesis that the ability of a machine to conduct convincing dialogues with human beings can serve as at least a sufficient criterion of AGI. We argue that this very ability should be accepted also as a necessary condition of AGI, and we provide a description of the nature of human dialogue in particular and of human language in general against this background. We then argue that it is for mathematical reasons impossible to program a machine in such a way that it could master human dialogue behaviour in its full generality. This is (1) because there are no traditional explicitly designed mathematical models that could be used as a starting point for creating such programs; and (2) because even the sorts of automated models generated by using machine learning, which have been used successfully in areas such as machine translation, cannot be extended to cope with human dialogue. If this is so, then we can conclude that a Turing machine also cannot possess AGI, because it fails to fulfil a necessary condition thereof. At the same time, however, we acknowledge the potential of Turing machines to master dialogue behaviour in highly restricted contexts, where what is called “narrow” AI can still be of considerable utility.

In “On Drawing Lines on a Map” (1995), I suggested that the different ways we have of drawing lines on maps open up a new perspective on ontology, resting on a distinction between two sorts of boundaries: fiat and bona fide. “Fiat” means, roughly: human-demarcation-induced. “Bona fide” means, again roughly: a boundary constituted by some real physical discontinuity. I presented a general typology of boundaries based on this opposition and showed how it generates a corresponding typology of the different sorts of objects which boundaries determine or demarcate. In this paper, I describe how the theory of fiat boundaries has evolved since 1995, how it has been applied in areas such as property law and political geography, and how it is being used in contemporary work in formal and applied ontology, especially within the framework of Basic Formal Ontology.

Contains the following contributions: Ingvar Johansson: Ontologies and Concepts. Two Proposals Christopher Menzel: Reference Ontologies - Application Ontologies: Either/Or or Both/And? Luc Schneider: Foundational Ontologies and the Realist Bias Guenther Goerz, Kerstin Buecher, Bernd Ludwig, Frank-Peter Schweinberger, and Iman Thabet: Combining a Lexical Taxonomy with Domain Ontology in the Erlangen Dialogue System Vim Vandenberghe, Burkhard Schafer, John Kingston: Ontology Modelling in the Legal Domain - Realism Without Revisionism A Proposed Methodology for the Development of Application-Based Formal Ontologies Eric Little

The Plant Ontology (PO) is a community resource consisting of standardized terms, definitions, and logical relations describing plant structures and development stages, augmented by a large database of annotations from genomic and phenomic studies. This paper describes the structure of the ontology and the design principles we used in constructing PO terms for plant development stages. It also provides details of the methodology and rationale behind our revision and expansion of the PO to cover development stages for all plants, particularly the land plants (bryophytes through angiosperms). As a case study to illustrate the general approach, we examine variation in gene expression across embryo development stages in Arabidopsis and maize, demonstrating how the PO can be used to compare patterns of expression across stages and in developmentally different species. Although many genes appear to be active throughout embryo development, we identified a small set of uniquely expressed genes for each stage of embryo development and also between the two species. Evaluating the different sets of genes expressed during embryo development in Arabidopsis or maize may inform future studies of the divergent developmental pathways observed in monocotyledonous versus dicotyledonous species. The PO and its annotation databasemake plant data for any species more discoverable and accessible through common formats, thus providing support for applications in plant pathology, image analysis, and comparative development and evolution.

The ability to access and share data is key to optimizing and streamlining any industrial production process. Unfortunately, the manufacturing industry is stymied by a lack of interoperability among the systems by which data are produced and managed, and this is true both within and across organizations. In this paper, we describe our work to address this problem through the creation of a suite of modular ontologies representing the product life cycle and its successive phases, from design to end of life. We call this suite the Product Life Cycle (PLC) Ontologies. The suite extends proximately from The Common Core Ontologies (CCO) used widely in defense and intelligence circles, and ultimately from the Basic Formal Ontology (BFO), which serves as top level ontology for the CCO and for some 300 further ontologies. The PLC Ontologies were developed together, but they have been factored to cover particular domains such as design, manufacturing processes, and tools. We argue that these ontologies, when used together with standard public domain alignment and browsing tools created within the context of the Semantic Web, may offer a low-cost approach to solving increasingly costly problems of data management in the manufacturing industry.

Artificial intelligence (AI) research enjoyed an initial period of enthusiasm in the 1970s and 80s. But this enthusiasm was tempered by a long interlude of frustration when genuinely useful AI applications failed to be forthcoming. Today, we are experiencing once again a period of enthusiasm, fired above all by the successes of the technology of deep neural networks or deep machine learning. In this paper we draw attention to what we take to be serious problems underlying current views of artificial intelligence encouraged by these successes, especially in the domain of language processing. We then show an alternative approach to language-centric AI, in which we identify a role for philosophy.

The manufacturing industry is evolving rapidly, becoming more complex, more interconnected, and more geographically distributed. Competitive pressure and diversity of consumer demand are driving manufacturing companies to rely more and more on improved knowledge management practices. As a result, multiple software systems are being created to support the integration of data across the product life cycle. Unfortunately, these systems manifest a low degree of interoperability, and this creates problems, for instance when different enterprises or different branches of an enterprise interact. Common ontologies (consensus-based controlled vocabularies) have proved themselves in various domains as a valuable tool for solving such problems. In this paper, we present a consensus-based Additive Manufacturing Ontology (AMO) and illustrate its application in promoting re-usability in the field of dentistry product manufacturing.

La présente étude utilise les outils du domaine de la méréotopologie (la théorie des parts, ensembles et frontières) pour élaborer les implications de certaines analogies entre la 'psychologie écologique' de J.J.Gibson et la phénoménologie de Edmund Husserl. On présentera une théorie ontologique de frontières spatiales et des entités possédant une extension spatiale. S'en rapportant aux exemples de la sphère de géographie, on démontre qu'aussi bien les frontières que les entités à extension spatiale appartiennent à deux vastes catégories: des objets qui existent indépendamment de nos actes cognitifs (par exemple: la planète Terre, sa surface extérieure); et des objets qui existent seulement grâce à ces actes (par exemple: l'équateur, la Mer du Nord). Le champ visuel, lui aussi, peut être conçu comme un exemple d'une entité à extension spatiale qui est dépendante dans le sens proposé ci-dessus. L'étude présente suggère une extension de cette analogie en postulant les entités qui, pour les vrais jugements, seraient ce que le champ visuel est pour les actes de perception visuelle. Le champ de jugement est défini plus précisément comme une entité étendue complexe, contenant toutes les entités essentielles pour la vérité d'un (vrai) jugement donné. Il est possible de démontrer que le travail des linguistes cognitifs tel Talmy ou Langacker, si l'on l'interprète convenablement, produit une explication détaillée des structures des champs de jugement correspondant aux phrases de sortes différentes. Un genre nouveau de définition de la vérité en tant que correspondance pour les expressions de la langue naturelle peut donc être formulé sur cette base.

Human cognitive acts are directed towards entities of a wide range of different types. What follows is a new proposal for bringing order into this typological clutter. A categorial scheme for the objects of human cognition should be (1) critical and realistic. Cognitive subjects are liable to error, even to systematic error of the sort that is manifested by believers in the Pantheon of Olympian gods. Thus not all putative object-directed acts should be recognized as having objects of their own. Broadly, the objects towards which human cognition is directed should be parts of reality in a sense that is at least consistent with the truths of natural science. But such a scheme should also be (2) comprehensive: it should do justice to each sort of object on its own terms, and not attempt to eliminate objects of one sort in favour of objects of other, more favoured sorts. Linguistic and other forms of idealism, as well as Meinongian theories, which assign to each and every referring expression or intentional act an object tailored to fit, yield categorial schemes which fail to satisfy (1). Physicalistic and other forms of reductionism yield categorial schemes which fail to satisfy (2). What follows is a categorial scheme that is both critically realistic and comprehensive. Thus it enjoys some of the benefits of linguistic idealism and physicalism, without (or so it is hoped) the corresponding disadvantages of each.

A collection of papers presented at the First International Summer Institute in Cognitive Science, University at Buffalo, July 1994, including the following papers: ** Topological Foundations of Cognitive Science, Barry Smith ** The Bounds of Axiomatisation, Graham White ** Rethinking Boundaries, Wojciech Zelaniec ** Sheaf Mereology and Space Cognition, Jean Petitot ** A Mereotopological Definition of 'Point', Carola Eschenbach ** Discreteness, Finiteness, and the Structure of Topological Spaces, Christopher Habel ** Mass Reference and the Geometry of Solids, Almerindo E. Ojeda ** Defining a 'Doughnut' Made Difficult, N .M. Gotts ** A Theory of Spatial Regions with Indeterminate Boundaries, A.G. Cohn and N.M. Gotts ** Mereotopological Construction of Time from Events, Fabio Pianesi and Achille C. Varzi ** Computational Mereology: A Study of Part-of Relations for Multi-media Indexing, Wlodek Zadrozny and Michelle Kim

We describe a prototype ontology-driven information system (ODIS) that exploits what we call Portion of Reality (POR) representations. The system takes both sensor data and natural language text as inputs and composes on this basis logically structured POR assertions. The goal of our prototype is to represent both natural language and sensor data within a single framework that is able to support both axiomatic reasoning and computation. In addition, the framework should be capable of discovering and representing new kinds of situations and thematic roles, (e.g., roles such as agent, patient and instrument), based on new compositions of existing representations. We applied our prototype in an intelligence analysis use case to test the hypothesis that a framework of this sort can produce usefully structured information from combined natural language and sensor data inputs. We further tested our hypothesis by adding an enhanced US Air Force ontology framework to our ODIS in order to (1) process a collection of sensor data, intel reports, and mission plans; (2) build composite POR representations from these data; and (3) machine analyze the fused results to infer mission threats.

In a series of recent publications, orofacial researchers have debated the question of how ‘bruxism’ should be defined for the purposes of accurate diagnosis and reliable clinical research. Following the principles of realism-based ontology, we performed an analysis of the arguments involved. This revealed that the disagreements rested primarily on inconsistent use of terms, so that issues of ontology were thus obfuscated by shortfalls in terminology. In this paper, we demonstrate how bruxism terminology can be improved by paying attention to the relationships between (1) particulars and types, and (2) continuants and occurrents.

The Planteome project provides a suite of reference and species-specific ontologies for plants and annotations to genes and phenotypes. Ontologies serve as common standards for semantic integration of a large and growing corpus of plant genomics, phenomics and genetics data. The reference ontologies include the Plant Ontology, Plant Trait Ontology, and the Plant Experimental Conditions Ontology developed by the Planteome project, along with the Gene Ontology, Chemical Entities of Biological Interest, Phenotype and Attribute Ontology, and others. The project also provides access to species-specific Crop Ontologies developed by various plant breeding and research communities from around the world. We provide integrated data on plant traits, phenotypes, and gene function and expression from 95 plant taxa, annotated with reference ontology terms.

While representation learning techniques have shown great promise in application to a number of different NLP tasks, they have had little impact on the problem of ontology matching. Unlike past work that has focused on feature engineering, we present a novel representation learning approach that is tailored to the ontology matching task. Our approach is based on embedding ontological terms in a high-dimensional Euclidean space. This embedding is derived on the basis of a novel phrase retrofitting strategy through which semantic similarity information becomes inscribed onto fields of pre-trained word vectors. The resulting framework also incorporates a novel outlier detection mechanism based on a denoising autoencoder that is shown to improve performance. An ontology matching system derived using the proposed framework achieved an F-score of 94% on an alignment scenario involving the Adult Mouse Anatomical Dictionary and the Foundational Model of Anatomy ontology (FMA) as targets. This compares favorably with the best performing systems on the Ontology Alignment Evaluation Initiative anatomy challenge. We performed additional experiments on aligning FMA to NCI Thesaurus and to SNOMED CT based on a reference alignment extracted from the UMLS Metathesaurus. Our system obtained overall F-scores of 93.2% and 89.2% for these experiments, thus achieving state-of-the-art results.

The current industrial revolution is said to be driven by the digitization that exploits connected information across all aspects of manufacturing. Standards have been recognized as an important enabler. Ontology-based information standard may provide benefits not offered by current information standards. Although there have been ontologies developed in the industrial manufacturing domain, they have been fragmented and inconsistent, and little has received a standard status. With successes in developing coherent ontologies in the biological, biomedical, and financial domains, an effort called Industrial Ontologies Foundry (IOF) has been formed to pursue the same goal for the industrial manufacturing domain. However, developing a coherent ontology covering the entire industrial manufacturing domain has been known to be a mountainous challenge because of the multidisciplinary nature of manufacturing. To manage the scope and expectations, the IOF community kicked-off its effort with a proof-of-concept (POC) project. This paper describes the developments within the project. It also provides a brief update on the IOF organizational set up.

In the visual representation of ontologies, in particular of part-whole relationships, it is customary to use graph theory as the representational background. We claim here that the standard graph-based approach has a number of limitations, and we propose instead a new representation of part-whole structures for ontologies, and describe the results of experiments designed to show the effectiveness of this new proposal especially as concerns reduction of visual complexity. The proposal is developed to serve visualization of ontologies conformant to the Basic Formal Ontology. But it can be used also for more general applications, particularly in the biomedical domain.

Polish translation of "John Searle: From Speech Acts to Social Reality", We provide an overview of Searle's contributions to speech act theory and the ontology of social reality, focusing on his theory of constitutive rules. In early versions of this theory, Searle proposed that all such rules have the form 'X counts as Y in context C' formula – as for example when Barack Obama (X) counts as President of the United States (Y) in the context of US political affairs. Crucially, the X and the Y terms are here identical. A problem arises for this theory for cases involving 'free-standing Y terms', as for example in the case of money in a computerized bank account. Here there is no physical X to which a status function might be attached. We conclude by arguing that Searle's response to this problem creates difficulties for his naturalistic framework.

We discuss recent progress in the development of cognitive ontologies and summarize three challenges in the coordinated development and application of these resources. Challenge 1 is to adopt a standardized definition for cognitive processes. We describe three possibilities and recommend one that is consistent with the standard view in cognitive and biomedical sciences. Challenge 2 is harmonization. Gaps and conflicts in representation must be resolved so that these resources can be combined for mark-up and interpretation of multi-modal data. Finally, Challenge 3 is to test the utility of these resources for large-scale annotation of data, search and query, and knowledge discovery and integration. As term definitions are tested and revised, harmonization should enable coordinated updates across ontologies. However, the true test of these definitions will be in their community-wide adoption which will test whether they support valid inferences about psychological and neuroscientific data.

Vaccine research, as well as the development, testing, clinical trials, and commercial uses of vaccines involve complex processes with various biological data that include gene and protein expression, analysis of molecular and cellular interactions, study of tissue and whole body responses, and extensive epidemiological modeling. Although many data resources are available to meet different aspects of vaccine needs, it remains a challenge how we are to standardize vaccine annotation, integrate data about varied vaccine types and resources, and support advanced vaccine data analysis and inference. To address these problems, the community-based Vaccine Ontology (VO) has been developed through collaboration with vaccine researchers and many national and international centers and programs, including the National Center for Biomedical Ontology (NCBO), the Infectious Disease Ontology (IDO) Initiative, and the Ontology for Biomedical Investigations (OBI). VO utilizes the Basic Formal Ontology (BFO) as the top ontology and the Relation Ontology (RO) for definition of term relationships. VO is represented in the Web Ontology Language (OWL) and edited using the Protégé-OWL. Currently VO contains more than 2000 terms and relationships. VO emphasizes on classification of vaccines and vaccine components, vaccine quality and phenotypes, and host immune response to vaccines. These reflect different aspects of vaccine composition and biology and can thus be used to model individual vaccines. More than 200 licensed vaccines and many vaccine candidates in research or clinical trials have been modeled in VO. VO is being used for vaccine literature mining through collaboration with the National Center for Integrative Biomedical Informatics (NCIBI). Multiple VO applications will be presented.

The Cell Ontology (CL) is designed to provide a standardized representation of cell types for data annotation. Currently, the CL employs multiple is_a relations, defining cell types in terms of histological, functional, and lineage properties, and the majority of definitions are written with sufficient generality to hold across multiple species. This approach limits the CL’s utility for cross-species data integration. To address this problem, we developed a method for the ontological representation of cells and applied this method to develop a dendritic cell ontology (DC-CL). DC-CL subtypes are delineated on the basis of surface protein expression, systematically including both species-general and species-specific types and optimizing DC-CL for the analysis of flow cytometry data. This approach brings benefits in the form of increased accuracy, support for reasoning, and interoperability with other ontology resources. 104. Barry Smith, “Toward a Realistic Science of Environments”, Ecological Psychology, 2009, 21 (2), April-June, 121-130. Abstract: The perceptual psychologist J. J. Gibson embraces a radically externalistic view of mind and action. We have, for Gibson, not a Cartesian mind or soul, with its interior theater of contents and the consequent problem of explaining how this mind or soul and its psychological environment can succeed in grasping physical objects external to itself. Rather, we have a perceiving, acting organism, whose perceptions and actions are always already tuned to the parts and moments, the things and surfaces, of its external environment. We describe how on this basis Gibson sought to develop a realist science of environments which will be ‘consistent with physics, mechanics, optics, acoustics, and chemistry’.