3rd Conference on Language, Data and Knowledge (LDK)
LDK 2021 - the 3rd Conference on Language, Data and Knowledge - took place in Zaragoza. It was held as a hybrid event. It was supported by the NexusLinguarum COST Action CA18209 - European network for Web-centred linguistic data science, and participation to the conference was free of charge. NexusLinguarum also supported the proceedings as an open access publication by OASICS. The conference received additional support from the European H2020 project Prêt-à-LLOD and from the Universidad Zaragoza.
The main conference took place on September 2nd and September 3rd. It was located at the beautiful venue of the Paraninfo Building of the University of Zaragoza, which is hosting not only its ceremonial hall but also a library and a museum of natural sciences.
Workshops
Six pre-conference tutorials (LiLa LDK 2021 Tutorial, DBpedia Tutorial at LDK 2021) and four workshops (1st Workshop on Sentiment Analysis & Linguistic Linked Data, 4th shared task for Translation Inference Across Dictionaries (TIAD 2021), [2nd International Workshop on Artificial Intelligence for Historical Image and Visual Cultural Artefacts Enrichment (AI4HI-2021)])https://chia.acdh.oeaw.ac.at/ai4hi-2021-workshop/), Multisensory Data & Knowledge Workshop) took place on the 1st of September and two meetings within a post conference W3C Day took place on the 4th of September (W3C LD4LT Community Group and (W3C OntoLex Community Group).
The pre- and post- conference events were located at Salón de Actos of the Ada Byron Building of the School of Engineering and Architecture of the University of Zaragoza
Invited Speakers
Three speakers accepted the invitation to give a keynote talk. Mathieu Lafourcade of the University of Montpellier gave a presentation on “Collecting and inferring common sense knowledge: the JeuxDeMots project”. Sara Tonelli of the Fondazione Bruno Kessler gave a talk on “A Smell is Worth a Thousand Words: Olfactory Information Extraction and Semantic Processing in a Multilingual Perspective”. And Mikel Forcada of the University of Alicante spoke about “Free/open-source machine translation for the low-resource languages of Spain”.
Program
71 Papers were submitted and single-blinded reviewed by a committee of 69 reviewers. 38 Papers accepted (54%). A number of papers were shortlisted for best paper awards:
- Bridging the gap between Ontology and Lexicon via Class-specific Association Rules Mined from a Loosely-Parallel Text-Data Corpus by Basil Ell, Mohammad Fazleh Elahi, and Philipp Cimiano
- A review and cluster analysis of German polarity resources for sentiment analysis by Bettina M. J. Kern, Andreas Baumann, Thomas E. Kolb, Katharina Sekanina, Klaus Hofmann, Tanja Wissik, and Julia Neidhardt
- An ontology for CoNLL-RDF: Formal data structures for TSV formats in language technology by Christian Chiarcos, Maxim Ionov, Luis Glaser, and Christian Fäth
- Towards Learning Terminological Concept Systems from Multilingual Natural Language Text by Lennart Wachowiak, Christian Lang, Barbara Heinisch, and Dagmar Gromann
Best papers awards and a best poster award were finally selected:
- Best paper award: Bridging the gap between Ontology and Lexicon via Class-specific Association Rules Mined from a Loosely-Parallel Text-Data Corpus by Basil Ell, Mohammad Fazleh Elahi, and Philipp Cimiano
- Best student paper award:Towards Learning Terminological Concept Systems from Multilingual Natural Language Text by Lennart Wachowiak, Christian Lang, Barbara Heinisch, and Dagmar Gromann
- Best poster award: A data augmentation approach for sign-language-to-text translation in-the-wild Fabrizio Nunnari, Cristina España-Bonet, and Eleftherios Avramidis
The main conference was organized in 8 sessions:
- Language Data – Corpus Annotations and Mining
- Use Cases and Applications in Language, Data and Knowledge
- Knowledge Graphs – Generation and Bias
- Crazy New Ideas
- Multilingual Information Extraction and Language Learning
- Language Data and Applications – Semantic Enrichment and Discourse Analysis
- Under-Resourced Languages and Enriching Lexical Data
- Knowledge Graphs, Word Embeddings and Natural Language Generation
- A (hybrid) poster session was organized at the end of the first day, just before the social events, consisting in a touristic tour of Zaragoza and the conference dinner.
We initially counted on 258 registered participants, and the total number of unique participants was 212, 40 of them joining the conference physically in Zaragoza.
The conference attracted a high number of participants, more than doubling the number of participants of the former editions. We think that this is to a large part due to the support by NexusLinguarum, allowing a free registration to the event.