Editorial: Special Issue on Semantic Web and Ontology Design for Cultural Heritage

After an ex-ante impact assessment of a European Collaborative Cloud for Cultural Heritage (ECCCH) [2], a Horizon Europe call¹ is running for Culture, Creativity, and Inclusive Society that asks for a cloud enhancing “the ability of cultural heritage actors to interact across disciplinary, institutional, sectorial and political boundaries and cooperate effectively in advancing research on cultural heritage and in developing innovative solutions for the discovery, recovery, conservation, digitalisation and valorisation of digital, digitised and digitisable cultural heritage objects.” Interestingly, the ex-ante report and this call specify that such a cloud should “enable semantic representation of multiple data types (various incarnations of 2D and 3D media, video, text), stored in federated repositories according to FAIR principles, and encoding data provenance.” Other European initiatives multiply efforts towards Web infrastructures based on, or including more and more formally defined knowledge: Europeana has started building the common European Data Space for Cultural Heritage (the Data Space),² FAIRcore4EOSC³ and FAIR-IMPACT⁴ are developing coordinated solutions for greater and more harmonised use of semantic artefacts, leading to semantic interoperability within and between disciplines, including the numerous and varied fields of Cultural Heritage. At the system architecture level as well, SOLID⁵ is based on the Semantic Web standards.

This growing interest in scaling up robust semantic solutions is at the service of “Humanist Knowledge” building, in line with Reference [4], i.e., where knowledge representation and the Semantic Web provide the elements of socio-technical infrastructures in which humanists and professionals of Cultural Heritage can pursue specialist work but with bridges between different research areas, professional sectors, and disciplines. On the Ontology Design side, while the CIDOC CRM ecosystem is now well recognized as the heart of interoperability in the field of cultural heritage, there is no slowing down when it comes to the formal and necessarily careful definition of the very diverse concepts used in this multi-disciplinary field. This special issue, as well as the associated series of Semantic Web and Ontology Design for Cultural Heritage (SWODCH) workshops,⁶ show that the multidisciplinary scientific community working towards this aim is dynamic and productive.

Twenty-four submissions were reviewed by a large pool of specialists, acknowledged and thanked at the end of this Editorial. Ten of these submissions were accepted and published in this Special Issue after two or three rounds of reviews and revisions. These can be classified in three categories according to their main area or domain of contribution: the quest for the best possible ontological model for various domains and applications (five articles), the development of applications based on ontologies and Semantic Web technologies (two articles), and the design of models and tools dedicated to building, or to analysing knowledge graphs, i.e., Linked Data (three articles).

The articles in this special issue demonstrate the rise of concrete applications of ontologies, Semantic Web technologies and Linked Data principles for CH, together with better tools to support the development of such applications. As already noted in a previous special issue [1], ontology design and data modelling in this domain are guided by the re-usability principles of FAIR data, which recommend that (meta)data should also be “FAIR-ified,” without, however, paying less attention to soundness and quality, ensured to a large extent by the adoption of standard ontologies, such as CIDOC-CRM. More generally, the Semantic Web and Linked Data drive a great deal of research in CH as their aims (e.g., decentralising, sharing, opening, breaking data-silos, collaborating) are perfectly aligned with the particular needs of this domain.

Regarding future research in this area, one of the most compelling topics, to which we would like to turn the attention of scholars, concerns the challenges that recent developments in sub-symbolic artificial intelligence, such as machine learning and statistical applications, pose to formal knowledge modelling and, more generally, to the area of knowledge representation and reasoning. According to the recently published “Manifesto from Dagstuhl Perspectives Workshop 22282: Current and Future Challenges in Knowledge Representation and Reasoning” [3], among these challenges are undoubtedly those related to (i) the design of intelligent systems based on commonsense reasoning, that is, “the ability to make effective use of ordinary, everyday, experiential knowledge in achieving ordinary, everyday, practical goals,” whose impact spreads from natural language understanding, to constructing visual interpretations, planning, and interacting with the real world; (ii) the development of innovative and reliable techniques and tools to support knowledge acquisition and management, “that is, for the formulation of suitable domain knowledge by knowledge engineers and domain experts”; (iii) the design of hybrid systems in which machine learning techniques (e.g., deep neural networks, Generative and Large Language Models) are integrated with the symbolic representation of knowledge and associated reasoning techniques; (iv) the possibility of giving human-intelligible explanations of predictions or recommendations made by models developed using machine learning. To reap the benefits of both formal knowledge representation and statistical machine learning, DH scholars should not ignore such challenges and the research questions they imply. We expect that the maturity gained in this field, which we directly witnessed in the SWODCH workshops and the related follow-up initiatives, will lead to original and effective solutions, tailored to the specific needs of CH applications, but also transferable to other domains with similar characteristics.

The following list includes the names of the reviewers (in alphabetical order) who participated in the review process of the manuscripts submitted to the special issue. Their effort and excellent work have been fundamental for the publication of the high-quality scientific contributions composing the present JOCCH volume.

Trond Aalberg, Alessandro Adamou, Alessio Antonini, Salvatore Arcidiacono, Luigi Asprino, Valentina Bartalesi, Dimitris Bilidas, George Bruseker, Francesco Corcoglioniti, Marilena Daquino, Oyvind Eide, Stefano Faralli, Catherine Faron-Zucker, Achille Felicetti, Géraud Fokou, Richard Gartner, Manolis Georgatsoulis, Xavier Granier, Anais Guillem, Anas Fahad Khan, Mikko Koho, Konstantinos Kotis, Davide Lanti, Sylvain Laubé, Cristiano Longo, Marinucci Ludovica, Ludovica Marinucci, Yannis Marketakis, Claudio Masolo, Daniele Metilli, Andrea Giovanni Nuzzolese, Laura Pandolfo, Maria Angela Pellegrino, Lighton Phiri, Davide Picca, Emilio M. Sanfilippo, Daniele F. Santamaria, Michalis Sfakakis, Daria Spampinato, Sofia Stamou, Maria Rosaria Stufano Melone, Maria Theodoridou, Konstantin Todorov, Francesca Tomasi, Cássia Trojahn, Christos Tryfonopoulos, Valentina Vassallo, Costas Vassilakis, Fabio Vitali, Andreas Vlachidis, Dan Vodislav.