The interoperability of the data models that describe authority data is an important requirement for exchanging the data of Memory Organizations in the World Wide Web. In this paper we focus on the semantic interoperability between archival and library authorities. Initially we mapped EAC-CPF to MARC21 Authority standard and then we mapped EAC-CPF to CIDOC Conceptual Reference Model, which is an ontology for the cultural heritage. The study of the models and the methodologies for the development of the proposed mappings revealed the differences of information needs each model addresses.
Research metadata is difficult to annotate yet critical for research replication and data re-usability. A language called MEDFORD has been previously designed to make the process of writing metadata approachable for non-software focused researchers, but we have identified several flaws that make it difficult to thoroughly and accurately describe research metadata. We propose an extension to the MEDFORD metadata language to support relationships between metadata objects, as well as to reference other MEDFORD files, such that objects can be defined in a singular file and re-used throughout multiple files.
Technology reflects the values of its developers, which reinforces the need for ethical and responsible development in digital solutions. Awareness of gender equality in Information Science is reflected in United Nations Resolution 62/10 and the Sustainable Development Goals (SDGs), which promote gender equality. The aim of this research is to promote advances in the representation of women and girls in the Bibliodata Network, based on the use of sources such as the Virtual International Authority File (VIAF) and Wikidata for the semantic enrichment of metadata, as well as the importance of inclusive terminology to achieve this end. Although the study focuses mainly on enriching the records of people who identify as female, it is hoped in the future that the methodology employed will be adaptable for all genders. The expected results aim to promote significant advances in the representation of women and girls in the Bibliodata Network, improve the quality and accuracy of metadata, and create more robust information networks, with an ongoing commitment to equity and inclusion. Currently, the main challenge of the research is the migration and reorganization of data within the network to align with updated standards, ensuring data preservation and preventing potential loss.
This paper describes the integration of cross-disciplinary data from Solid Earth and Marine sciences through a metadata-driven approach within the Geo-INQUIRE project. It highlights the importance of Research Infrastructures (RIs) in managing and providing open access to scientific data, addressing challenges faced by distributed RIs in ensuring multidisciplinary data access. It also highlights the implementation of Virtual Access and Physical or TransNational Access in line with European Union regulations. Additionally, it discusses the successful implementation of a proof of concept between EPOS and EMSO RIs, showcasing the potential of cross-infrastructure metadata-driven integration for supporting comprehensive and multidisciplinary research.
Open datasets are often exposed with insufficient metadata, making difficult to end users the task of identifying those that better fit their needs. One way to overcome this weaknesses is to guarantee compliance of data to the FAIR principles, in particular where the use of ontologies is a key aspect for proving richer metadata schemes. This paper proposes an ontology network, DATA-FW, that aims at representing rich metadata to assist in dataset usage. It exploits different features that are required to meet the user's needs, which have been divided into four distinct components: Core Metadata Component, Structure Component, Usage Component, and Quality Component. Each Component reuses existing and known vocabularies and serves a specific purpose.
The National Library of Greece (NLG) adopted new cataloguing standards, namely IFLA LRM and RDA, two international standards which adhere to linked data principles. Even though newly created metadata is linked data compliant, the legacy data is not. Considering the objective of publishing the data of the NLG to the linked data cloud, interoperability in terms of technology, standards, and language is essential. Towards this goal, workflows have changed, corrections and enrichments were made, new system functionalities were developed, and new tools were used. This paper presents the implemented approach.
This paper explores the convergence of Open Data initiatives, Linked Data technologies, ontological knowledge representation, and Large Language Models (LLMs) in generative Artificial Intelligence (AI). It examines how these complementary approaches can be integrated to create more powerful, flexible, and context-aware knowledge systems. The paper provides an overview of the open data landscape, the Semantic Web and Linked Data vision, ontologies and knowledge organization systems, and recent advances in LLMs. It then discusses how these technologies can be synergistically combined to enable next-generation knowledge systems that leverage both structured knowledge and natural language understanding. Potential applications in areas such as scientific research, government transparency, and intelligent information retrieval are discussed. The paper also addresses key challenges including scalability, data quality, ethical considerations, and the need for explainable AI. A strategic roadmap for realizing this integration is proposed, emphasizing collaboration between academia, industry, and government. While significant technical and ethical challenges remain, the convergence of these technologies has the potential to fundamentally transform how we interact with and derive insights from information, enabling more intelligent and context-aware knowledge systems to address complex real-world problems.
Historical documents are vital for preserving cultural information. To access their content, OCR and HTR technologies transcribe them into text. Challenges arise with handwritten texts due to varied writing styles. This study develops two corpora to train models for 19th-century Greek texts. The first dataset includes printed texts from the "Ellinomnimon" archive, and the second comprises hand-written documents from the Lasithi Demogerontia archives. Moreover, using the Transkribus platform, two models were iteratively refined, enhancing their ability to transcribe Greek historical documents from 1800 to 1870.
A thematic path is a tool for the enjoyment of cultural heritage that allows the connection of cultural objects of different nature through the identification of common themes. This study aims to develop the Cultural Thematic Path (CTP) Ontology, an ontology for the creation and structuring of cultural thematic paths, in order to support the linked data publishing and managing for the enjoyment of cultural heritage in the GLAM (Galleries, Libraries, Archives, and Museums) domain. Despite the literature revealing that numerous ontologies deal with managing and describing cultural resources related to the GLAM domain, the absence of a schema capable of fully depicting and representing thematic paths is highlighted. In this work, all the phases of ontology development are presented, from design assumptions to the ontology requirements specifications, from the implementation activity to the description of the ontology's defined entities and properties.
The Coordination of Bibliographic Services (Cobib/Ibict) has modernized its systems, including the National Collective Catalog of Serial Publications (CCN), in the initiative called Pinakes, with the aim of updating the data generated over decades. This work aims to present the correspondence between the entities of the IFLA Library Reference Model (IFLA LRM) and the classes of the Pinakes domain. From the IFLA LRM diagram, the main entities were identified and defined, mapping them to specific classes of the Pinakes Model, categorized and numbered. The relationships and their cardinalities were established, highlighting the importance of inverse properties for the organization of the data. This mapping provides a robust initial framework for the Pinakes Model, which is essential for the organization and retrieval of bibliographic information. In addition, this work lays a solid foundation for future expansions and refinements of the model, facilitating the development of a comprehensive ontology for the Pinakes catalog.
Each language has its own body of emotion vocabulary. It is also possible to map emotion terms of one language to similar terms in another language. However, the mere existence of similar terms across languages does not guarantee that emotion concepts being translated are identical. Emotion terms can evoke different feelings in different individuals and often embody the societal and cultural norms of a linguistic community. Hence, conveying emotions across languages poses a unique challenge in translation. This paper explores the role of knowledge organisation systems and synonym rings for enabling better understanding and encoding of emotionally weighted material in translations. The study uses synonym rings to present emotion vocabulary across Sinhala and English and employs contextual evidence and literary text translations to analyse and evaluate the findings. The research identifies and utilises three pillars of inquiry: (1) Cognition, (2) linguistics, and (3) socio-cultural factors to explore the ways in which individuals perceive and express emotions across Sinhala and English.
The ontology presented in this paper completes and extends the Academic Events Ontology (AEON), which reuses several ontologies from the OBO foundry, especially the information artifact ontology (IAO). The present work formally represents the education-related event domain. The challenge is to describe the knowledge underlying the organization of educational meetings, academic and professional, and model it using semantic technologies. To capture and exchange structured knowledge in this domain, an ontology should address the organization of educational meetings in all steps, along with the associated scientific evidence base. It describes event formats, venues, calls for papers, the audience, information on submission, and fees. In addition, it addresses organizations and people involved in the process. The ontology was created using the NeOn methodology and OBO foundry guidelines for ontology development. Several domain experts also provided their expertise. The ontology is available on GitHub and licensed via an open-source license.
The role of ontologies in facilitating search capabilities within large collections of data is critical; the integration and analysis of diverse data sources becomes feasible as ontologies frame the data conceptually and provide a common understanding of terms and their relationships- the lack of ontological and conceptual support entailing the opposite effect. Along with documents and data lost in the vast-ness of available yet disparate data sources, numerous scientific papers and published research remain undiscovered due to poor linking to their respective scientific domain and investigation method(s) described in them. Within the scope of this study is to retrieve existing Wikidata method codes for 3 disciplines: psychology, neuroscience and cultural heritage, and analyse them, with the purpose of identifying gaps in the usage of hierarchical levels or codes, and examining whether they are currently capable of sufficiently describing the methodological domains in question, while also pertaining to a suitable level of specificity in order for the related data to be efficiently and effectively queried and retrieved. The findings revealed several issues regarding the discoverability and semantic search capabilities to retrieve scientific literature papers on research (or investigation) methods and tooling for the in-word disciplines. In this light, a proposed methodology to alleviate the current situation is drafted, introducing the utilisation of technological means, such as LLMs, to assist in identifying orphan categories of methods or tools and, by benchmarking against basic existing ontologies (e.g., FrameNet or other related Linked Open Vocabularies), to enrich the hierarchical structure of current representation practices in this regard.
Historical data are sparse and vague making their collection and analysis difficult. To come up with a methodology for handling such data, in this paper we elaborate on this problem by considering the requirements and challenges for building a Knowledge Graph that contains data about both modern and historical earthquakes. We discuss the requirements and challenges, and we report the main observations concerning uncertainties by analyzing the documentation of ancient earthquakes that hit the island of Crete. Afterwards, we analyze representation approaches and related ontologies and then we present a modeling approach that can tackle the requirements. We showcase the feasibility of the approach by implementing it using RDF and we share publicly the result.
Our world is becoming more interconnected everyday and there is an increasing demand for smart solutions on diverse data management. Ontology-based systems enhance data interoperability and analysis across various domains and platforms. In healthcare, they enable the integration of patient records, medical histories, and treatment plans across various providers, thereby improving care quality and patient outcomes. In the tourism industry, they facilitate the effective management of travel information, lodgings and activity options, leading to more personalized and efficient travel experiences. A unified Health Tourism ontology is presented in this manuscript aiming to enhance the overall quality of life by offering personalized health and wellness travel experiences that cater to individual needs and preferences. This paper details the design, implementation, and utilization of the proposed ontology, exploring its application in a Health Tourism application and its potential for broader use in other domains requiring structured data management and interoperability.
In the present article, we explore the morphological semantics of Modern Greek (MG) and how these are expressed in ontological terms in the MMoOn ontology. First, we define the two core aspects of meaning, grammatical and lexical, pointing out that there are also other more abstract types of meaning such as derivational that pertain to affixes or confixes. Most prominently, we explore the transition of features of morphemic components to the output derivational structures, a process most commonly known as percolation theory. In order to ensure the consistency of the process, we specify an automated way of this transfer as well as the types of the transmitted features for each different case. To this end, we leverage OWL expressive language, defining appropriate classes, object properties and chain properties in the Protégé ontology editor. Afterwards, we confirm that the syntax works well by checking out the generated inferences. In the research to come, more investigation on morphemic semantics of MG is to be done touching upon other aspects of it.
The National Library of Greece (NLG) transliterates Greek authors' and corporate bodies' names to Latin characters, and integrates the transliterated text into authority records. Submitting data to the Virtual Authority File (VIAF) revealed that transliterations help cluster the Greek version of names under the proper cluster. To further enhance shareability and matching potential of the NLG data, it was decided to enrich name authority records with transliterations under two standards: ALA-LC Romanization tables, and ELOT743 (equivalent to ISO843). The NLG collaborated with the International Hellenic University, Department of Information and Electronic Engineering, and the Open Knowledge Foundation Greece to build a transliterator tool. This paper presents the development of the tool, along with the challenges in applying the rules that each standard determines. The tool was evaluated against a Gold Dataset created by the NLG staff.
After a winding development path full of challenges throughout history, thanks to the intervention of technology, public libraries have evolved in the last decades from traditional libraries to true ecosystems of operational services grafted on the current needs of users. Being perceived rather as conservative institutions, in order to remain relevant, public libraries have been forced to innovate, especially in terms of service-related aspects. Gradually, under the pressure of users, they have integrated various applications and technologies which today allow for a better connection with the audience and a better solution to the specific needs of library service beneficiaries. This paper explores the different understandings of the concept of "renewable knowledge" in the context of public libraries, positioning these institutions as essential spaces for the creation, transfer and long-term capitalization of this knowledge. Through a systematic clarification of terminology, the study analyzes in depth the dynamics of generating and preserving renewable knowledge, emphasizing their role as a universal public good that encourages creativity, co-creation and transformative learning. This research provides a future-oriented perspective on understanding and modeling "renewable knowledge" in libraries and the impact of this transformation on the knowledge assets of learning communities. The author investigates and critically evaluates how the generation of knowledge - understood as the result of interpersonal interactions - makes libraries become nodal centers for conservation, valorization and reuse of renewable knowledge elements. By analyzing the complex process of aggregating renewable knowledge and its distinctive features compared to other knowledge typologies, the present research attempts to clarify how foundational knowledge can be transformed into renewable knowledge. Current work highlights the potential of libraries to serve as critical facilitators in this process of transforming collective knowledge into renewable knowledge assets, leveraging the unique and privileged position of libraries within communities. This article investigates the interest of library and information science professionals in identifying reusable facets of knowledge from the value-added perspective of modern technology-based library services. Using a mix consisting of a Questionnaire and a Structured Interview, this paper explores the role of public libraries as catalysts in the complex process of renewable knowledge development. Highlighting the dynamic and adaptive nature of the acquisition and dissemination of renewable knowledge, this article examines the potential of public libraries to creatively engage with communities, integrate technology, and propagate knowledge in a coherent and sustainable manner. The findings attest to the idea that public libraries have the ability to adapt to the changing information landscape, strengthening their democratic role in the contemporary knowledge society and contributing to the development of intelligent, inclusive and connected communities. In supporting the research results, the author offers as an example a concrete case of application of the concept of renewable knowledge in libraries, carried out within the Horizon project SHIFT: MetamorphoSis of cultural Heritage Into augmented hypermedia assets For enhanced accessibiliTy and inclusion. Being a use study based on the recent practice of libraries in Romania, the practical example refers to the empowerment of pre-existing digital stories with new values based on elements of renewable knowledge. This use case demonstrates once more that the integration of advanced digital technologies can substantially revive the preservation and accessibility of cultural heritage, making it accessible even for users belonging to vulnerable groups.
Music ontology is a framework that is used to publish structured data related to music and became available on the Semantic Web through data interfaces. This process is articulated in this article through the analysis of the Musical Greek Audiovisual Collections (M.EL.O.S.) project and its ontology. Specifically, the project involves the integration of three different types of music collections, gathered on a common platform, Reasonable Graph (RG), while retaining the autonomy of the participating institutions. The suggested ontology was initially based on the international conceptual models FRBR, FRBRoo, FRAD and FRSAD and their WEMI (Work, Expression, Manifestation, Item) ontologies, while their structure and role in the formation of the musical ontology of the M.EL.O.S. project was also analysed. Being evident that the original ontology could not cover the description needs of the music ontology, the already existing music ontology "MUSIC ONTOLOGY" was additionally utilized, adding the established type of entities "Agent" (Person, Organization, Family) and "Subject" (Concept, Place, Event, Object, Genre and Subject Chain). Thus, additional fields specialized metadata descriptions appeared. The Reasonable Graph platform is the one that eventually supported music ontology, data interconnection, and the ability to access and research music data with a focus on interoperability and the detailed descriptions of the music materials. Thus, through visualisation examples of the final product, the connection between metadata, entities and conceptual models is explained, while the processes and findings of the analyses within the M.EL.O.S. project opened new paths in data interconnection and the importance of abundance in it. Thus, this paper goes beyond the common presentation of basic elements of music ontology and focuses on its development, the steps and logic it followed, the specializations, and its enrichment, providing a basis for future references and approaches to organizing and managing complex musical information. It also covers the key outcomes of the project and its ontology, which contribute to the development of an appropriate music ontology as a result of the M.EL.O.S. project outcomes. This includes processes such as discussions about project and ontology needs, the automated extraction of music content from participating institutions, the crowdsourcing procedures, and the development of collaborative operations. These efforts were implemented on the Reasonable Graph system, alongside the transfer of metadata and digital/digitized documents.
This paper presents JobHive, a recommender system based on knowledge graphs to provide improved recommendations by aligning candidate resumes with job requirements, considering both explicit and implicit skills. By integrating semantic similarity computations, the system ensures comprehensive match quality for job seekers and employers. The matching algorithm calculates a similarity score between job offers and resumes by comparing skills, experience, and inferred skills. It uses a Transformer-based Sequential Denoising Auto-Encoder (TSDAE) for contextualized understanding, which generates comprehensive representations of entities to improve semantic similarity assessments. Additionally, the algorithm uses a knowledge graph to understand connections between entities, allowing it to find the best matches by considering both direct and indirect relationships. The evaluation of the matching algorithm for JobHive demonstrated its effectiveness in ranking resumes according to job offers. Tested with 40 job offers and 240 resumes, the algorithm achieved high relevance scores, indicating it closely matched manual rankings.
