2020, Elias, Mirette, Tavakoli, Mohammadreza, Lohmann, Steffen, Kismihok, Gabor, Auer, Sören, Gurreiro, Tiago, Nicolau, Hugo, Moffatt, Karyn

Open Educational Resources are becoming a significant source of learning that are widely used for various educational purposes and levels. Learners have diverse backgrounds and needs, especially when it comes to learners with accessibility requirements. Persons with disabilities have significantly lower employment rates partly due to the lack of access to education and vocational rehabilitation and training. It is not surprising therefore, that providing high quality OERs that facilitate the self-development towards specific jobs and skills on the labor market in the light of special preferences of learners with disabilities is difficult. In this paper, we introduce a personalized OER recommeder system that considers skills, occupations, and accessibility properties of learners to retrieve the most adequate and high-quality OERs. This is done by: 1) describing the profile of learners with disabilities, 2) collecting and analysing more than 1,500 OERs, 3) filtering OERs based on their accessibility features and predicted quality, and 4) providing personalised OER recommendations for learners according to their accessibility needs. As a result, the OERs retrieved by our method proved to satisfy more accessibility checks than other OERs. Moreover, we evaluated our results with five experts in educating people with visual and cognitive impairments. The evaluation showed that our recommendations are potentially helpful for learners with accessibility needs.

2021, Dessì, Danilo, Käser, Tanja, Marras, Mirko, Popescu, Elvira, Sack, Harald, Dessì, Danilo, Käser, Tanja, Marras, Mirko, Popescu, Elvira, Sack, Harald

The design and delivering of platforms for online education is fostering increasingly intense research. Scaling up education online brings new emerging needs related with hardly manageable classes, overwhelming content alternatives, and academic dishonesty while interacting remotely, as examples. However, with the impressive progress of the data mining and machine learning fields, combined with the large amounts of learning-related data and high-performance computing, it has been possible to gain a deeper understanding of the nature of learning and teaching online. Methods at the analytical and algorithmic levels are constantly being developed and hybrid approaches are receiving an increasing attention. Recent methods are analyzing not only the online traces left by students a posteriori, but also the extent to which this data can be turned into actionable insights and models, to support the above needs in a computationally efficient, adaptive and timely way. In this paper, we present relevant open challenges lying at the intersection between the machine learning and educational communities, that need to be addressed to further develop the field of intelligent systems for online education. Several areas of research in this field are identified, such as data availability and sharing, time-wise and multi-modal data modelling, generalizability, fairness, explainability, interpretability, privacy, and ethics behind models delivered for supporting education. Practical challenges and recommendations for possible research directions are provided for each of them, paving the way for future advances in this field.

2022, Sohmen, Lucia, Rack, Fabian, Heuveline, Vincent, Bisheh, Nina

Die zunehmende Verfügbarkeit von Forschungsdaten eröffnet Forschenden neue Möglichkeiten, mit von Dritten erstellten Forschungsdaten zu arbeiten. Dieser Beitrag befasst sich mit der Frage, welche rechtlichen Rahmenbedingungen gelten, wenn diese nachgenutzten Forschungsdaten öffentlich verfügbar gemacht werden sollen. Im Speziellen geht der Artikel dabei auf Bildersuchmaschinen und das Veröffentlichen von Bildkorpora ein. Dabei wird dargestellt, dass es bei der öffentlichen Zugänglichmachung von unübersichtlichen Bildmengen keine hundertprozentige Sicherheit geben kann. Durch bestimmte Abwägungen und technische Mittel kann sich dieser aber angenähert werden.

2021, Bruns, Oleksandra, Tietz, Tabea, Vafaie, Mahsa, Dessì, Danilo, Sack, Harald, Lopes, Carla Teixeira, Ribeiro, Cristina, Niccolucci, Franco, Rodrigues, Irene, Freire, Nuno

Archival records are essential sources of information for historians and digital humanists to understand history. For modern information systems they are often analysed and integrated into Knowledge Graphs for better access, interoperability and re-use. However, due to restrictions of the representation of RDF predicates temporal data within archival records is a challenge to model. This position paper explains requirements for modeling temporal data in archival records based on running research projects in which archival records are analysed and integrated in Knowledge Graphs for research and exploration.

2021, Vafaie, Mahsa, Bruns, Oleksandra, Pilz, Nastasja, Dessì, Danilo, Sack, Harald, Sumikawa, Yasunobu, Ikejiri, Ryohei, Doucet, Antoine, Pfanzelter, Eva, Hasanuzzaman, Mohammed, Dias, Gaël, Milligan, Ian, Jatowt, Adam

An increasing number of archival institutions aim to provide public access to historical documents. Ontologies have been designed, developed and utilised to model the archival description of historical documents and to enable interoperability between different information sources. However, due to the heterogeneous nature of archives and archival systems, current ontologies for the representation of archival content do not always cover all existing structural organisation forms equallywell. After briefly contextualising the heterogeneity in the hierarchical structure of German archives, this paper describes and evaluates differences between two archival ontologies, ArDO and RiC-O, and their approaches to modelling hierarchy levels and archive dynamics.

2021, Stegmüller, Johannes, Bauer-Marquart, Fabian, Meuschke, Norman, Ruas, Terry, Schubotz, Moritz, Gipp, Bela, Zhang, Chengzhi, Mayr, Philipp, Lu, Wie, Zhang, Yi

Identifying cross-language plagiarism is challenging, especially for distant language pairs and sense-for-sense translations. We introduce the new multilingual retrieval model Cross-Language Ontology-Based Similarity Analysis (CL-OSA) for this task. CL-OSA represents documents as entity vectors obtained from the open knowledge graph Wikidata. Opposed to other methods, CL-OSA does not require computationally expensive machine translation, nor pre-training using comparable or parallel corpora. It reliably disambiguates homonyms and scales to allow its application toWebscale document collections. We show that CL-OSA outperforms state-of-the-art methods for retrieving candidate documents from five large, topically diverse test corpora that include distant language pairs like Japanese-English. For identifying cross-language plagiarism at the character level, CL-OSA primarily improves the detection of sense-for-sense translations. For these challenging cases, CL-OSA’s performance in terms of the well-established PlagDet score exceeds that of the best competitor by more than factor two. The code and data of our study are openly available.

2022, Hartmann, Thomas, Heuveline, Vincent, Bisheh, Nina

Ein Schlüsselfaktor für die Zugänglichkeit, Nachnutzbarkeit und Interoperabilität von Forschungsdaten1 ist deren urheberrechtlicher Status. In den Geistes- und Sozialwissenschaften unterliegen Forschungsdaten (z. Bsp. Texte in Form von Interviews oder sonstiger Literatur) in den meisten Fällen dem Urheberschutz. Anders ist die Situation in den Naturwissenschaften. Nicht immer, aber häufig bleiben Forschungsdaten aus diesen Fächern urheberrechtsfrei. Dies begründet der Beitrag an einem typischen Beispiel aus dem Forschungsdatenzentrum für Molekulare Materialforschung (SDC MoMaF).2 Welche rechtlichen Handlungsempfehlungen sich für das Forschungsdatenmanagement (FDM) daraus ergeben, wird am Beitragsende dargestellt.

2021, Tan, Mary Ann, Tietz, Tabea, Bruns, Oleksandra, Oppenlaender, Jonas, Dessì, Danilo, Harald, Sack, Sumikawa, Yasunobu, Ikejiri, Ryohei, Doucet, Antoine, Pfanzelter, Eva, Hasanuzzaman, Mohammed, Dias, Gaël, Milligan, Ian, Jatowt, Adam

Under the German government’s initiative “NEUSTART Kultur”, the German Digital Library or Deutsche Digitale Bibliothek (DDB) is undergoing improvements to enhance user-experience. As an initial step, emphasis is placed on creating a knowledge graph from the bibliographic record collection of the DDB. This paper discusses the challenges facing the DDB in terms of retrieval and the solutions in addressing them. In particular, limitations of the current data model or ontology to represent bibliographic metadata is analyzed through concrete examples. This study presents the complete ontological mapping from DDB-Europeana Data Model (DDB-EDM) to FaBiO, and a prototype of the DDB-KG made available as a SPARQL endpoint. The suitabiliy of the target ontology is demonstrated with SPARQL queries formulated from competency questions.

2021, Tan, Mary Ann, Tietz, Tabea, Bruns, Oleksandra, Oppenlaender, Jonas, Dessì, Danilo, Sack, Harald, Seneviratne, Oshani, Pesquita, Catia, Sequeda, Juan, Etcheverry, Lorena

Cultural heritage portals have the goal of providing users with seamless access to all their resources. This paper introduces initial efforts for a user-oriented restructuring of the German Digital Library (DDB). At present, cultural heritage objects (CHOs) in the DDB are modeled using an extended version of the Europeana Data Model (DDBEDM), which negatively impacts usability and exploration. These challenges can be addressed by leveraging ontologies, and building a knowledge graph from the DDB's voluminous collection. Towards this goal, an alignment of bibliographic metadata from DDB-EDM to FRBR-Aligned Bibliographic Ontology (FaBiO) is presented.

2022, Landwehr, Matthias, Schneider, Gabriel, Hofmann, Stefan, Razum, Matthias, Soltau, Kerstin, Heuveline, Vincent, Bisheh, Nina

Die Universität Konstanz deckt ihren Bedarf nach einem institutionellen Forschungsdatenrepositorium mit der von FIZ Karlsruhe angebotenen Lösung „RADAR Local“. Als Alternative zu einer Eigenentwicklung wurde das Datenrepositorium als hybrides Modell mit Repositorien-Software und Archivierung auf lokaler Infrastruktur implementiert. Dabei stellt FIZ Karlsruhe die etablierte Repositorien-Software RADAR zur Verfügung, wartet und betreibt sie aus der Ferne und passt sie nach Kundenwunsch an. Um die parallele Installation und Pflege der RADAR-Software auf mehreren lokalen Instanzen effizient bewältigen zu können, hat FIZ Karlsruhe vorab den Automatisierungsgrad der betroffenen Prozesse in der Software-Entwicklung, in der Systemkonfiguration und im Deployment erhöht. Dies wurde durch den Einsatz von Container-Virtualisierung wie Docker und Docker Swarm sowie mit Orchestrierungswerkzeugen wie Ansible erreicht. Der zeitliche Aufwand und der personelle Ressourcenbedarf reduzieren sich dadurch für die Universität Konstanz und als Ergebnis erhält sie ein gepflegtes Repositorium auf dem aktuellen Stand der Technik. Gleichzeitig erfordert diese Betriebsvariante eine intensive Auseinandersetzung mit dem jeweiligen Geschäftsmodell und den technischen Rahmenbedingungen des Anbieters, eine genaue Kostenkalkulation sowie möglicherweise Kompromisse oder Abstriche bei individuellen Wünschen.