Ebook: Modular Ontologies
Modularity, in its different shapes and forms, remains one of the central research topics in ontology engineering, still catching up with 40 years of related research in software engineering. The workshops on Modular Ontologies (WoMO) bring together researchers from different disciplines who study the problem of modularity in ontologies at a fundamental level, develop design tools for distributed ontology engineering and apply modularity in different use cases and application scenarios.The contributions in this volume are of interest to researchers, students and practitioners interested in Semantic Web ontologies, their languages and tools and specifically, to research groups working on ontology modularization and integration problems and corresponding tools. They should also be of interest to the broader communities of knowledge representation and reasoning, information integration, description logics and ontology languages, distributed systems and to practitioners from Semantic Web and other emerging application domains for ontologies such as life sciences, robotics, e-business and ambient intelligence.
Modularity, in its different shapes and forms, remains one of the central research topics in ontology engineering. An already huge and ever-growing number of ontologies can be found on the Internet. These ontologies vary dramatically in terms of subject matter, levels of specification and detail, (intended) purpose and (actual) application. Not only is it difficult to determine and define interrelations between such distributed ontologies, with contributions from different sources covering different parts of a single domain, it is also challenging to reconcile (jointly inconsistent) ontologies each of which covering a particular aspect of the same domain, to extract relevant parts of an ontology, or to re-combine, in various ways, small ontologies in order to form new, larger ones.
Still catching up with 40 years of related research in software engineering, ontological modularity is envisaged to allow mechanisms for easy and flexible reuse, generalisation, structuring, maintenance, design patterns, and comprehension. Applied to ontology engineering, modularity is central not only to reduce the complexity of understanding ontologies, but also in maintaining, querying and reasoning over modules. Distinctions between modules can be drawn on the basis of structural, semantic, or functional aspects, which can also be applied to compositions of ontologies or to indicate links between ontologies.
In particular, reuse and sharing of information and resources across ontologies depend on purpose-dependent, logically versatile criteria. Such purposes include ‘tight’ logical integration of different ontologies (wholly or in part), ‘loose’ association and information exchange, the detection of overlapping parts, traversing through different ontologies, alignment of vocabularies, module extraction possibly respecting privacy concerns and hiding of information, etc. Another important aspect of modularity in ontologies is the problem of evaluating the quality of single modules or of the achieved overall modularization of an ontology. Again, such evaluations can be based on various (semantic or syntactic) criteria and employ a variety of statistical/heuristic or logical methods.
Recent research on ontology modularity has produced substantial results and approaches towards foundations of modularity, techniques of modularisation and modular developments, distributed reasoning and empirical evaluations, providing a foundation for further research and development. Since the beginning of the WoMO workshop series, there has been growing interest in the modularisation of ontologies, modular development of ontologies, and information exchange across different modular ontologies. In real life, however, integration problems are still mostly tackled in an ad-hoc manner, with no clear notion of what to expect from the resulting ontological structure. The adoption of ad hoc methods for combining ontologies often leads to unintended consequences, even if the individual ontologies to be integrated are widely tested and understood.
Topics covered by WoMO include, but are not limited to:
What is Modularity?
- kinds of modules and their properties
- modules vs. contexts
- design patterns
- granularity of representation
- robustness properties such as conservativity etc.
- modular ontology languages (e.g., DDL, [Escr ]-Connections, P-DL)
- reconciling inconsistencies across modules
- formal structuring of modules
- distributed reasoning
- modularisation and module extraction
- sharing and reusing, linking and importing
- evaluation of modularisation approaches
- complexity of reasoning
- reasoners or implemented systems
- modularity in the Semantic Web
- Life Sciences, Ambient Intelligence, e-Business, robotics, etc.
- selective information sharing, hiding, and privacy
- collaborative ontology development
The WoMO 2010 workshop follows a series of successful events that have been an excellent venue for practitioners and researchers to discuss latest work and current problems. It is intended to consolidate cutting-edge approaches that tackle the problem of ontological modularity and bring together researchers from different disciplines who study the problem of modularity in ontologies at a fundamental level, develop design tools for distributed ontology engineering, and apply modularity in different use cases and application scenarios. Previous editions of WoMO were:
WoMO 2006 The 1st workshop on modular ontologies, co-located with ISWC 2006, Athens, Georgia, USA. Invited speakers were Alex Borgida (Rutgers) and Frank Wolter (Liverpool). The web site is at http://www.cild.iastate.edu/ events/womo.html and the proceedings at http://sunsite.informatik.rwth-aachen.de/Publications/CEUR-WS/Vol-232
WoMO 2007 The 2nd workshop, co-located with K-CAP 2007, Whistler BC, Canada. The invited speaker was Ken Barker (Texas at Austin). The web site is at http://webrum.uni-mannheim.de/math/lski/WoMO07 and the proceedings at http://sunsite.informatik.rwth-aachen.de/Publications/CEUR-WS/Vol-315
WoRM 2008 The 3rd workshop in the series, co-located with ESWC 2008, Tenerife, Spain, entitled ‘Ontologies: Reasoning and Modularity’ had a special emphasis on reasoning methods. The web site is at http://dkm.fbk.eu/worm08 and the proceedings at http://sunsite.informatik.rwth-aachen.de/Publications/CEUR-WS/Vol-348
Overview of Contributions
The invited speakers for WoMO 2010 addressed rather general and somewhat unconventional issues related to modularity in ontologies.
SIMON COLTON, in his talk “Towards Ontology Use, Re-use and Abuse in a Computational Creativity Collective”, described the Computational Creativity Collective project, and in particular, possibilities for using techniques such as automated theory formation and analogical reasoning for ontological engineering, and how ontological modularity will have a role in more creative reasoning via the collective.
In the second invited talk, entitled “Ontology Modularity, Information Flow, and Interaction-Situated Semantics”, MARCO SCHORLEMMER sketched the landscape of modularity and distributed reasoning in IT systems and discussed the interesting topic of the role of channel theory and information flow, as introduced by Jon Barwise and Jerry Seligman, with respect to ontological modularity and meaning negotiation.
The contributed papers of WoMO 2010 covered a wide range of topics, from mathematical foundations, to empirical evaluations of modularity, extraction and merging of modules, as well as applications in areas such as complex event processing and robotics.
The paper by DEL VESCOVO, PARSIA, SATTLER, AND SCHNEIDER, “The modular structure of an ontology: an empirical study”, gives an empirical analysis of the problem and the feasibility of extracting all modules of real-life ontologies, based on the concept of locality-based modules. They in particular discuss to what extent revealing the modular structure of an ontology in this way allows to obtain information about its topicality, connectedness, structure, superfluous parts, or agreement between actual and intended modelling.
HUSSAIN AND ABIDI present a method for extracting ontology modules in “Extracting and Merging Contextualized Ontology Modules”. The extraction is based on a user-defined selection, which indicates those parts and constructs of an ontology that are relevant for the module extraction. The method is particularly applicable to the health-care domain.
In “A Metric Suite for Evaluating Cohesion and Coupling in Modular Ontologies”, ENSAN AND DU give another perspective on the evaluation problem in modular ontologies. They introduce semantic metrics for measuring cohesion and coupling in ontologies. The calculation of these metrics is based on analysing relativeness and dependencies between local symbols and between local and external symbols of modules. In particular, the authors use the “Kitchenham” metric validation framework in order to examine their presented metrics.
POKRYWCZYŃSKI AND MALCOLM continue a tradition to use the abstract framework of institutions in ontology engineering. In “Towards a Functional Approach to Modular Ontologies using Institutions”, they propose a general notion of framework that, irrespective of the formalisms being used, allows a global language into which both an ontology language and a query language can be translated. They then study generalisations of notions of robustness and conservativity in the heterogeneous case and their role for modularity.
In “Introducing ontology best practices and design patterns into robotics: USAREnv”, RANDELLI AND NARDI present an ontology for unstructured environments in the context of urban search and rescue activities. Among other information, the ontology specifies disaster causes, site types, and localisations. The authors further analyse the use of ontology design patterns and the re-use of existing ontologies to provide an agile implementation and the integration into different contexts.
The development of ontologies in the field of complex event processing can benefit particularly from modularity, as TEYMOURIAN AND PASCHKE argue in “Modular Upper-Level Ontologies for Semantic Complex Event Processing”. They distinguish different characteristics and dependencies of complex events and introduce several upper level ontology modules to specify these characteristics respectively. These upper level ontology modules can then provide a guideline for developing individual event ontologies.
Contributors to this volume
RAZA ABIDI is Professor of Computer Science and Director of Health Informatics at the Faculty of Computer Science, Dalhousie University. His research focuses on semantic web, web services and healthcare knowledge management for clinical decision support. He leads the NICHE Research Group that comprises around 20 inter-disciplinary researchers. He has published over 150 research articles and supervised over 50 graduate students.
SIMON COLTON is a Senior Lecturer in the Department of Computing at Imperial College London. He is an AI researcher who leads the Computational Creativity group http://www.doc.ic.ac.uk/ccg, where the focus of study is on building software which takes on some of the creative responsibility in arts and science projects, with applications to graphic design, discovery in pure mathematics, video game design and the visual arts.
GÖKHAN COSKUN is a Research Assistant in the Corporate Semantic Web working group at the Freie Universität Berlin (FUB), Germany. His research interests comprise network structure of ontologies and aspect-oriented ontology engineering. He is working on his doctoral thesis investigating on structural analysis and structure-based modularisation of ontologies.
BERNARDO CUENCA GRAU is a Royal Society University Research Fellow at the University of Oxford. His current research activities are focused on new reasoning services needed to support key tasks in Ontology Engineering, such as ontology merging, design, maintenance and integration. He has been involved in the design of the OWL 2 extension of the Web Ontology Language. He authored a number of papers on logical foundations for modular ontologies and module extraction at major conferences such as IJCAI, KR and WWW.
CHIARA DEL VESCOVO is a PhD Student at the University of Manchester, United Kingdom, in the Information Management Group. She graduated in Mathematics at the University Roma 3, Italy. After graduation, she worked in Research & Development department of CM Sistemi, a private informatics company. She is currently working on her PhD project, focused on modularity for ontology comprehension.
WEICHANG DU is a professor at the Faculty of Computer Science, University of New Brunswick, Canada. He obtained his PhD in Computer Science from University of Victoria, Canada, in 1991. His research interests include Semantic Web, Knowledge Engineering, Software Engineering, Programming Languages, Internet Computing, and Mobile Computing.
FAEZEH ENSAN joined the Faculty of Computer Science at the University of New Brunswick, Canada, as a PhD student in May 2006. Before joining UNB, she spent two years completing her Master of Science at the Ferdowsi University of Mashhad, after receiving her bachelor of Computer Engineering from the Faculty of Engineering, University of Tehran in 2004. Her research is focused on aspects of knowledge engineering, Semantic Web, ontology development, modular ontologies and belief revision.
JOANA HOIS is a Research Assistant in the Research Center on Spatial Cognition (SFB/TR 8) at the University of Bremen, Germany. She is a developer of the spatial module for the ‘Generalized Upper Model’, a linguistic ontology, and works on combining formal models of space with spatial language. Her current research activities are focused on modular ontologies of space in different domains and applications as well as on combining ontologies with different kinds of uncertainties.
SAJJAD HUSSAIN is an active research member in the NICHE Research Group, Faculty of Computer Science, Dalhousie University. He is doing his PhD in the area of Extracting and Merging Contextualised Modular Ontologies under the supervision of Raza Abidi. His research interests include (i) Ontology Modularisation, (ii) Ontology Reconciliation and Merging, and (iii) Ontology Debugging. He is involved in a number of research projects at NICHE and acknowledged as a developer in the developers team of the Euler reasoning engine.
OLIVER KUTZ is a Postdoctoral Research Fellow in the Research Center on Spatial Cognition (SFB/TR 8) at the University of Bremen, Germany. He published widely on spatial, modal and description logics, on counterpart semantics, as well as on structuring and heterogeneity in ontologies. He co-designed the logic [Sscr ][Rscr ][Oscr ][Iscr ][Qscr ] underlying the web ontology language OWL 2 as well as the [Escr ]-connections technique that is used in modular ontology design, and is a founding member of ‘The International Association for Ontology and its Applications’ (IAOA) established in 2009.
GRANT MALCOLM is a Lecturer in the Computer Science Department of the University of Liverpool, UK, where he is a member of the Logic and Computation Group. His major research interest lies in the field of algebraic specification. More specifically, his work addresses topics in the areas of hidden algebra, concurrent systems, computer virology, ontologies, algebraic semiotics, and modelling biological systems.
DANIELE NARDI is Full Professor at Sapienza Università di Roma, Dipartimento Informatica e Sistemistica. Education: Electronic Engineering Degree, Politecnico Torino, 1981, Master Computer and System Engineering, Sapienza Univ. Roma, 1984. Research interests: Artificial Intelligence, Cognitive Robotics, Search and Rescue Robotics. Recipient of “IJCAI-91 Publisher's Priz”, “Intelligenza Artificiale 1993”. Trustee of RoboCup Federation. Head of the research laboratory “Cognitive Robot Teams”.
BIJAN PARSIA is a Lecturer in the School of Computer Science at the University of Manchester, UK, where he is a member of the Information Management Group (IMG). He has published on many aspects of ontology engineering using description logics including on explanation, modularity, reasoning optimization, visualization, and representing uncertainty.
ADRIAN PASCHKE is Full Professor at the Freie Universität Berlin (FUB) and head of the AG Corporate Semantic Web (AG-CSW). He is research director at the Centre for Information Technology Transfer (CITT) GmbH, director of RuleML Inc., Canada, and vice director of the Semantics Technologies Institute Berlin (STI Berlin). He is steering-committee chair of the RuleML Web Rule Standardisation Initiative, co-chair of the Reaction RuleML technical group, founding member of the Event Processing Technology Society (EPTS) and chair of the EPTS Reference Architecture working group, voting member of OMG, and member of several W3C groups such as the W3C Rule Interchange Format (W3C RIF) working group where he is editor of several W3C Semantic Web standard specifications.
DANIEL POKRYWCZYŃSKI is a PhD student in the Computer Science Department of the University of Liverpool. In 2005, he has obtained an M.A. in philosophy from Nicolaus Copernicus University in Toruń, Poland. His current research interests are focused on category theory, the theory of institutions, and description logics and ontologies.
GABRIELE RANDELLI is a PhD Candidate at Dipartimento di Informatica e Sistemistica, Sapienza University of Rome, and he is a member of the “Cognitive Robot Teams” laboratory. His main research areas are human-robot interaction through symbolic representations and high-level user interfaces. He realised USAREnv, an ontology that models unstructured environments for search and rescue robotic activities.
ULI SATTLER is a professor in the Information Management Group within the School of Computer Science of the University of Manchester. She works in logic-based knowledge representation, investigates standard and novel reasoning problems, and designs algorithms for their usage in applications. Together with Ian Horrocks and others, she has developed the [Sscr ][Hscr ][Iscr ][Qscr ] family of description logics underlying OWL and OWL 2. Together with various colleagues, she has published several papers on the concept of inference-preserving modules, their computation, and usage.
THOMAS SCHNEIDER is a Research Associate in the Information Management group in the School of Computer Science at the University of Manchester. His PhD was on the computational complexity of hybrid logics, supervised by Martin Mundhenk from the University of Jena. He is now working on the EPSRC funded project “Composing and decomposing ontologies: a logic-based approach”, and has published a number of papers on how to use logic-based approaches to modularity to support various tasks in ontology engineering. He has developed the locality-based module extractor that is currently available, e.g., through the OWL API.
MARCO SCHORLEMMER received the licenciate and doctorate degrees in Informatics from the Technical University of Catalonia. He is currently a tenured scientist at CSIC's Artificial Intelligence Research Institute in Barcelona. He is an author of over fifty publications in the fields of formal specification and automated theorem proving, diagrammatic representation and reasoning, distributed knowledge coordination, and semantic interoperability of ontologies.
KIA TEYMOURIAN is a Research Assistant in the group of Networked Information Systems at the Freie Universität Berlin (FUB), Germany. His current research activities are focused on knowledge-based complex event processing as well as large-scale semanticenabled distributed information systems. He is pursuing a PhD focusing on the area of semantic rule-based event processing.
Efficiently extracting a module from a given ontology that captures all the ontology's knowledge about a set of specified terms is a well-understood task. This task can be based, for instance, on locality-based modules.
In contrast, extracting all modules of an ontology is computationally difficult because there can be exponentially many. However, it is reasonable to assume that, by revealing the modular structure of an ontology, we can obtain information about its topicality, connectedness, structure, superfluous parts, or agreement between actual and intended modeling. Furthermore, incremental reasoning makes use of a number of, although not all possible, modules of an ontology.
We report on experiments to estimate the number of modules of real-life ontologies. We also evaluate the modular structure of ontologies that we succeeded to fully modularise. In that evaluation, we look at the number and sizes of the modules, as well as the relation between module size and number and size of signatures that lead to the module. Chances are that the understanding we report about small ontologies can be applied to all ontologies.
Ontology module extraction, from a large ontology, leads to the generation of a specialized knowledge model that is pertinent to specific problems. Existing ontology module extraction methods tend to either render a too generalized or a too restricted ontology module that at times does not capture the entire semantics of the source ontology. We present an ontology module extraction method that extracts a contextualized ontology module whilst extending the semantics of the extracted concepts and their relationships in the ontology module. Our approach features the following tenets (i) identifying the user-selected concepts that are pertinent for the problem-context at hand; (ii) extracting the user-selected concepts, their roles and their individuals; and (iii) extracting other concepts, roles and individuals that are structurally-connected with the user-selected concepts. We apply our ontology module extraction method in the Healthcare domain, and demonstrate (a) extraction of ontology modules from three prostate cancer pathway ontologies; and then (b) merging of extracted ontology modules to generate a comprehensive therapeutic work-flow knowledge for prostate cancer care management.
In this paper, we present a set of semantic metrics for measuring cohesion and coupling in modular ontologies. For defining these metrics, we consider the semantic of modular ontologies instead of their syntax. Based on the semantic-based definitions, both explicitly asserted knowledge and the implied knowledge that is derived from the explicitly represented knowledge are considered for ontology evaluation. We validate the introduced metrics based on the Kitchenham metric validation framework.
We propose a functional view of ontologies that emphasises their role in determining answers to queries, irrespective of the formalism in which they are written. A notion of framework is introduced that captures the situation of a global language into which both an ontology language and a query language can be translated, in an abstract way. We then generalise existing notions of robustness from the literature, and relate these to interpolation properties that support modularisation of ontologies.
Systems with knowledge representation and reasoning functionality are quite common within the robotics community. Nevertheless, most of the proposed architectures are ad-hoc implementations, which lack of modularity, standardization, and that cannot be reused or shared among different users. We fill that the Semantic Web effort over the past years in promoting standard representations and ontology best practices should be adopted by the robotic community, to foster the design of more effective and reusable systems, adapt for non expert users in everyday activities. In this paper, we investigate how to integrate ontology best practices and standard representations into the robotic system design, modelling an ontology for a real application field: urban search and rescue robotics. We also discuss the benefits of this methodology for robotic systems, as well as presenting some effective design guidelines.
Event-driven systems are highly depending on the quality of detection and processing of events. Many of complex real-world events cannot be processed by the existing event processing systems because they are too complex to be understood and processed by the systems. Complex events can be inferred from raw primitive events based on their incoming sequence, their syntax and semantics. Usage of ontological background knowledge about events and their relationship to other non-event concepts can improve the quality of event processing. In this paper, we discuss the profits and problems of ontology developing and usage in the area of event processing and propose modular upper-level ontologies for semantic enabled complex event processing. We discuss, why the modularity is needed and how these ontologies should be modular build up based on ontology engineering aspects to be able to address scalability, expressiveness and reuse.