Tag Me If You Can! Semantic Annotation of Biodiversity Metadata with the QEMP Corpus and the BiodivTagger

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 4557–4564 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 4557 Tag Me If You Can! Semantic Annotation of Biodiversity Metadata with the QEMP Corpus and the BiodivTagger Felicitas L¨offler1 , Nora Abdelmageed1 2 , Samira Babalou1 , Pawandeep Kaur1 , Birgitta K¨onig-Ries1 2 3 1Heinz Nixdorf Chair for Distributed Information Systems, Friedrich Schiller University Jena, Germany 2Michael Stifel Center for Data-Driven and Simulation Science, Jena, Germany 3German Center for Integrative Biodiversity Research (iDiv), Halle-Jena-Leipzig, Germany {felicitas.loeffler, nora.abdelmageed, samira.babalou, pawandeep.kaur, birgitta.koenig-ries}@uni-jena.de Abstract Dataset Retrieval is gaining importance due to a large amount of research data and the great demand for reusing scientific data. Dataset Retrieval is mostly based on metadata, structured information about the primary data. Enriching these metadata with semantic annotations based on Linked Open Data (LOD) enables datasets, publications and authors to be connected and expands the search on semantically related terms. In this work, we introduce the BiodivTagger, an ontology-based Information Extraction pipeline, developed for metadata from biodiversity research. The system recognizes biological, physical and chemical processes, environmental terms, data parameters and phenotypes as well as materials and chemical compounds and links them to concepts in dedicated ontologies. To evaluate our pipeline, we created a gold standard of 50 metadata files (QEMP corpus) selected from five different data repositories in biodiversity research. To the best of our knowledge, this is the first annotated metadata corpus for biodiversity research data. The results reveal a mixed picture. While materials and data parameters are properly matched to ontological concepts in most cases, some ontological issues occurred for processes and environmental terms. Keywords: Semantic Annotation, Ontology-Based Information Extraction, Gold Standard, Metadata, Biodiversity Research, Life Sciences 1. Introduction Metadata are an important source in Dataset Retrieval (Sir- iJodha Khalsa, 2018) as they contain compressed informa- tion about data collection objects, geographic location, au- thor and temporal expressions in a structured and machine- readable form. Driven by research fields such as Semantic Publishing (Shadbolt et al., 2006) and initiatives such as the FAIR principles (Wilkinson et al., 2016), metadata are increasingly semantically enriched with Uniform Resource Identifiers (URI) based on Linked Open Data (LOD) (Heath and Bizer, 2011) to foster interoperability of information about authors, papers and datasets. Scholars also benefit from semantically enriched metadata in the retrieval pro- cess as the search results can be expanded on related infor- mation such as synonyms (L¨offler et al., 2017). One domain that requires such improved retrieval tech- niques and ontological enhancement is biodiversity re- search, a discipline dealing with the variety of species, the genetic diversity and the diversity of functions, interactions and ecosystems 1. Numerous text mining applications have been developed to detect named entities such as species, persons and geographic locations (Naderi et al., 2011), (Cunningham et al., 2013). However, based on our pre- vious research (L¨offler et al., 2020), we figured that further entity types are important to be considered, when search- ing for …

Content truncated to 500 words