Interconnecting the Semantic Web with the discipline of Philosophy
Ioanna Kyvernitou, first year DAH student in English at NUI Galway, surveys a range of knowledge representation technologies which allow effective linking among networks of data in the domain of Philosophy.
What are the benefits of applying digital humanities techniques to Philosophy and how do these applications affect the discipline? How can knowledge representation technologies be used to organise data deriving from philosophical resources? In recent years, the rise of collaborative projects on the Web and the interdisciplinary work that has been accomplished between target domain scholars and computer scientists is significant in terms of disseminating and making accessible to Internet users digital resources. More specifically, by using technologies of the Semantic Web, semantically annotated and structured data can be shared, reused and processed automatically or manually, revealing possible new relationships among pieces of data.
What is the Semantic Web?
The term ‘Semantic Web’ refers to W3C’s vision of the Web of linked data to ‘enable people to create data stores on the Web, build vocabularies, and write rules for handling data’ (W3.org) with the aim to enable a better working cooperation of computers and people (Berners-Lee et al.). Thus, the scope of the Semantic Web is to enable computers to process more easily and intelligently information on the web. In order to create a Web of data, data on the Web must be available in a ‘standard format (RDF), reachable and manageable by Semantic Web tools’. Also ‘relationships among data should be made available’ (W3.org) (Fig.1).
Figure 1: The semantic Web layers. Source: http://www.cs.vu.nl/~guus/talks/04-owl-brisbane/all.htm
What is an ontology?
The term ‘ontology’, used here, does not refer to the philosophical notion of ‘ontology’ but as defined in the field of Computer Science; it is a term that refers to a formal data structure. Specifically, ‘an ontology is a formal, explicit specification of a shared conceptualization’ (Hyvönen 62). ‘Explicit’ because all elements of an ontology are explicitly defined. ‘Formal’ since the ontology specification is given in a language with a formal syntax and semantics, which results in machine executable and machine interpretable ontology descriptions. The term, ‘shared’ describes the fact that an ‘ontology is representing consensual knowledge that has been agreed on by a group of people, typically as a result of a social process’ (Staab and Studer VIII – IX). Therefore, the development of ontologies helps to share common understanding of the structure of information, and to reuse and analyse domain knowledge (protege.stanford.edu).
Ontological Engineering and Philosophy: State of the art
In the discipline of Philosophy, implementations of ontology engineering have been made in order to ‘assist humans and automatic agents in understanding the contents of the domain (especially in terms of properties, relations, and subsumption/inheritance relationships which hold between the domain’s types) and to allow data generated in one project to be interoperable with others’ (Buckner, Niepert and Allen 208).
As Buckner, Niepert and Allen also mention, little work has been completed on ontologies of ideas. This is because ‘most ontologies focus on more stable taxonomic structures (especially on types of physical objects or positions in social hierarchies), and few focus on the classification of abstract objects.’ And as they further note, ‘this is to be expected, as the classificatory structure of abstract entities is much more likely to be unstable, vague, and controversial’ (211).
Therefore, the development of ontologies of ideas can be a really demanding and complex process. As Garbacz also notes, ‘the philosophical research has not produced a homogeneous body of consistent knowledge. This field of study is all about controversies, interpretations, and viewpoints.’ Thus, he suggests that ‘if we want to maximise the scope of the philosophical research to be represented by an engineering ontology, we need to minimise the impact of particular philosophical assumptions we make building the ontology’ (6).
Below, some paradigms of ontological engineering implementations within the discipline of Philosophy are presented in order to demonstrate the state of art of this interdisciplinary research and work.
- Indiana Philosophy Ontology (InPhO) project
The InPhO ontology was developed as a support tool for the Stanford Encyclopedia of Philosophy (SEP), an online, open access, dynamic reference work, with the aim to be also deployed in other digital philosophy applications. It uses a ‘combination of automated methods and expert feedback to create a dynamic computational ontology for the discipline of philosophy’ (Buckner, Niepert and Allen 207) (Fig.2). It is published as an OWL ontology and the top-most layer ‘contains six main categories: Human, Idea, Nationality, Organization, Profession, and Publication’ (Garbacz 2).
Figure 2: ‘The InPhO layer cake: The SEP provides expert generated content at the bottom, and expert review at the top, while in between are layers of software and people with less expertise, who nevertheless make valuable contributions.’ Source: http://firstmonday.org/ojs/index.php/fm/article/view/2214/2023
- PhiloSURFical ontology
The PhiloSURFical pedagogical application was prototyped with Wittgenstein’s Tractatus Logico Philosophicus, a classical philosophical work of the twentieth century. It allows the navigation of a semantically enhanced version of the text. The PhiloSURFical ontology is data component of the PhiloSURFical application. It was ‘created to describe the philosophical domain at various levels of abstraction.’ As described in the project’s website (Pasin), the ontology enables users to ‘reorganize the same text according to the relevance of a single metadata,’ ‘query the knowledge base or other repositories in the Semantic Web, such as the DBpedia.’ As described (Pasin), ‘this is achieved by using simultaneously the knowledge encoded in the ontology, an initial knowledge base of resources and metadata built by a philosophy teacher, and the SPARQL query language to gather information from other sources in the Semantic Web.’
The ontology was developed with the use of Operational Conceptual Modelling Language (OCML), providing ‘import/export mechanisms from OCML to other languages, such as OWL and Ontolingua,’ which ‘ensure interoperability with current standards’ (Pasin).
Discovery Project is another example of ‘integrated project, bringing together both humanities scholars and IT specialists, stakeholders from both the academy and industry’. It ‘aims to harness the power of emerging Semantic Web technology for the purposes of research in philosophy’ (D’Iorio and Bartscherer 269,272).
Figure 3: Philospace, is ‘a network of personal desktop applications used to enrich the content of Philosource with semantic annotations’. Philosource, is ‘a federation of semantic digital libraries in the field of philosophy’ (Discovery-project.eu).
For the sake of the Discovery project, Philospace ontology was developed. Philospace is an ‘application based on Dbin, that runs on a local machine and is connected to other machines via a peer-to-peer network.’ It enables users to ‘add comments to documents (or parts of documents) and establish semantic relationships between philosophical concepts.’ This is achievable because in Philospace ‘knowledge is expressed using Semantic Web standard languages, fostering data interoperability between the two platforms and even with external applications’ (Discovery-project.eu).
The Wittgenstein ontology was developed in order to permit interlinked browsing of philosophical concepts and primary sources regarding Wittgenstein’s work found in the Wittgenstein Archives at the University of Bergen (WAB). It was implemented using OWL and it ‘received a significant boost from the EU-supported DISCOVERY project (2006-09, Smith 2007)’ (Dh2012.uni-hamburg.de). It was developed further within the framework of the NordForsk funded JNUVWAB project (2008-11), and it continued in the EU funded projects AGORA (2011-13) and DM2E (2012-14), and the Norwegian National Library funded project DIGITALE FULLTEKSTARKIV (DF) (2012-13) at the University of Bergen Library (Wab.uib.no). The Wittgenstein ontology ‘includes both classes for sources and for philosophical concepts as well as properties which relate these classes’ (Dh2012.uni-hamburg.de) (Fig.4) and its design is based in the Discovery project (Pichler and Zollner-Weber 702).
Figure 4: The structure of the Wittgenstein ontology (screenshot from Protégé). Source:http://dm2e.eu/files/Pichler_et_al_Sharing_and_debating_Wittgenstein.pdf
Ontology of Metaphysics project aims to ‘build a computational model of philosophical research with the help of the tools and methods available in knowledge representation.’ The the project aims to ‘represent philosophy as an activity (“philosophy-as-a-process”) and philosophy as its outcome (“philosophy-as-a-result”)’ (Metaontology.pl). It is also noted that the scope of the current project is ‘restricted to metaphysics, which is the branch of philosophy that influenced the development of the discipline of knowledge representation’. Moreover, ‘the resulting applied ontology is to be a component of a web portal that provides information on the philosophical research in Poland’ (Metaontology.pl).
Figure 5: OntOfOnt class hierarchy. Source: http://garbaczp.nazwa.pl/ftp/OntologyOfMetaphysics/Papers/PhilKR.pdf
OntOfOnt is an OWL ontology. Garbacz mentions that ‘as of the time of writing the ontology has 39 classes, which are related by 39 object properties.’ (Fig.5) He also notes that ‘the ontology’s axiomatisation attempts to employ the full expressive power available in OWL 2 DL, so its expressivity is ALCROIQ(D)’ (9). Lastly, he describes that ‘since OntOfOnt is not a foundational ontology, its upper level categories were “stitched” up to the categories of two foundational ontologies: CIDOC CRM and Information Artefact Ontology’ (9).
As shown above, the interconnection of the Semantic Web with the discipline of Philosophy by using knowledge representation technologies, makes possible inferencing and reasoning and thus it enables new methods to emerge in representing philosophical resources. It also enables users to semantically annotate data, effectively retrieve and analyse information within an efficiently structured web.
Berners-Lee, Tim, James Hendler, and Ora Lassila. ‘The Semantic Web’. Scientific American Cs.umd.edu., 2001. Web. 6 Nov. 2015.
Buckner, Cameron, Mathias Niepert, and Colin Allen. ‘From Encyclopedia To Ontology: Toward Dynamic Representation Of The Discipline Of Philosophy’. Synthese 182.2 (2010): 205-233. Web.
Dh2012.uni-hamburg.de,. ‘Towards Wittgenstein On The Semantic Web | Digital Humanities 2012′. N.p., 2015. Web. 6 Nov. 2015.
Discovery-project.eu,. ‘Discovery Project – The Discovery Project: Philosophy In The Digital Era?’.2015. Web. 6 Nov. 2015.
D’Iorio, Paolo & Bartscherer, Thomas (2008). Philosophy in an Evolving Web: Necessary Conditions, Web Technologies, and the Discovery Project. In Alois Pichler & Herbert Hrachovec (eds.), Philosophy of the Information Society: Proceedings of the 30th International Ludwig Wittgenstein-Symposium in Kirchberg, 2007. De Gruyter 261-274.
Garbacz, Pawel. Challenges For Ontological Engineering In The Humanities–A Case Study Of Philosophy. 2015. Web. 6 Nov. 2015.
Hyvönen, Eero. Publishing And Using Cultural Heritage Linked Data On The Semantic Web. San Rafael, Calif.: Morgan & Claypool Publishers, 2012. Print.
Metaontology.pl,. ‘Project Description | Ontology Of Metaphysics’., 2015. Web. 6 Nov. 2015.
Pasin, Michele. ‘Philosurfical’. Philosurfical.open.ac.uk., 2007. Web. 6 Nov. 2015.
Pichler, A., and A. Zollner-Weber. ‘Sharing And Debating Wittgenstein By Using An Ontology’. Literary and Linguistic Computing 28.4 (2013): 700-707. Web.
Protege.stanford.edu,. ‘What Is An Ontology And Why We Need It’. N.p., 2015. Web. 6 Nov. 2015.
Staab, Steffen, and Rudi Studer. Handbook On Ontologies. Berlin: Springer, 2009. Print.
W3.org,. ‘Data – W3C’. N.p., 2015. Web. 6 Nov. 2015.
W3.org,. ‘Semantic Web – W3C’. N.p., 2015. Web. 6 Nov. 2015.
Wab.uib.no,. N.p., 2015. Web. 6 Nov. 2015.