Ontology engineering in Digital humanities

Tags

,

Digital Humanities as a science field still has explicit ontology issues. With development and distribution of Digital Humanities the need for developing the methodology will only grow, increasing the attention on ontologies.

Ontology engineering methods in humanities have been the focus of systematic work for several years now in many projects. In previous years it served just as a formal representation of a set of concepts and their relationships within a single domain, because without semantic frameworks to tie it together, all missives of data are useless. As a result, for many domains there are multiple different scattered ontologies rather than one standard vocabulary. Nowadays the focus is changing, future tasks of humanities can involve sharing data across projects and repositories, that’s why the next step is to make these frameworks common.  

Next three papers show the examples of applying ontology engineering to Digital Humanities studies, implementing more universal ontological approach.

The aim of the paper on ‘Building a metrical ontology as a model to link digital poetic repertoires‘ is to model a basis for linking poetic repertoires. The problem of linking poetic repertoires is compounded not only by technical issues involved, but also conceptual problems, cultural differences. The goal of semantic description was to create an ontology, whose elements can be used in other poetic repertoires. There is no yet a conceptual model of ontology in a poetry domain, but this approach was applied to datasets of repertoires of different countries and traditions. This means that this project of a poetic ontology has a potential to become something more than a repository of datasets, but instead to create a semantic standardized structure to describe, analyze and work with the different poetic digital repertoires.

An ontology for 3D visualisation of cultural heritage” [2] proposes an ontology for defining 3D objects. This concept allows keeping data about 3D models, linking different 3D models, making the visualizations searchable, even re-using them. The interesting point in this ontology is the possibility to refine the ontology itself in the process of collaborative work. 

The Geolat project presented in “Annotating texts with ontologies, from geography to persons and events” [3] is an example of geographical ontology which is used for annotating place names in texts. The project is focused on Latin literature for the prototype phase, but with right ontological approach it might be expanded to include more works and other languages, or even other metadata such as people instead of toponyms.

Comparing these papers gives an insight into possible directions of ontology engineering development in Digital Humanities. Each of them represents an example of benefits derived from the use of a more universal ontological approach. While a model in [1] and a prototype in [3] have a potential to become a “common language” between different (poetical or geographical) resources by using semantic standardized structures; the proposal in [2] is not so detailed ontology and can’t be used out-of-the-box, but serves as an example of applying of Semantic Web principles to visualisation (instead of principles dictated by the traditions of museology).

The reconciliation of different ontologies may seem impossible, because of semantic heterogeneity problem in ontologies we already discussed in example of poetic terminological traditions, but the work of a team from Bar-Ilan University “Collaboratively maximizing inter-ontology agreement for controversial domains: A case study of Jewish cultural heritage” [4] shows that it is possible to construct a reliable large consensual ontology even for controversial domains with use of ontology matching algorithms.

I dare to assume that Digital Humanities field is moving in direction of development of the universal standards for ontology engineering. For example, it cannot be said that the goal of the Semantic Web movement is already attained, but standards proposed by W3C provided the World Wide Web with a common framework that allows data to be shared and reused without boundaries. Standardizing ontologies in the domain may also seem as quite audacious and unattainable goal, but it can be the next step in the development of Digital Humanities as a science.

 References:

[1] González-Blanco, Elena; Seláf, Levente; Del Rio Riande, María Gimena; Martínez Cantón, Clara Isabel; Martos Pérez, María Dolores (2014), Building a metrical ontology as a model to link digital poetic repertoires
[3] Lana, Maurizio; Ciotti, Fabio; Magro, Diego ; Peroni, Silvio; Tomasi, Francesca; Vitali, Fabio (2014), Annotating texts with ontologies, from geography to persons and events
Follow

Get every new post delivered to your Inbox.

Join 125 other followers