EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

RECODE - Relational Ecological COrpus for Data Extraction

Vasco V Branco iD, Lidia Pivovarova, Kari-E J Lintulaakso, Luís M Correia, Lenka Baranovičová iD, Iiris Lahin, Francisco Dias, David Filipe, Pedro Cardoso iD

Biodiversity Data Journal · 17.04.2026 · 14

DOI: 10.3897/bdj.14.e177365 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Biodiversity Data Journal, online erschienen am 17.04.2026. Verfasst von Vasco V Branco und 8 weiteren Autor:innen. Beteiligte Einrichtung: Finnish Museum of Natural History LUOMUS, University of Helsinki, Helsinki, Finland Finnish Museum of Natural History LUOMUS, University of Helsinki …. Die Arbeit nennt 22 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractSchlagwörterBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Ecology, conservation biology and related disciplines are inherently data-based, with the success of many research projects and initiatives (e.g., protected areas, monitoring plans etc.) being directly dependent on the availability of location and trait information on species and populations. Unfortunately, this data is often either non-existent or available only as unstructured text within publications, especially for megadiverse taxa such as many invertebrate orders. With the emergence of large language models, there have been many attempts to automatically parse such data in machine-readable formats with variable success, either using prompt engineering or training models fit-for-purpose through named entity recognition and relation extraction. Model training has proven more efficient for complex data relations, but it needs labelled corpora, i.e. curated training data containing examples of this information for models to statistically learn from. This is a time-consuming process and, to our knowledge, no standard datasets exist upon which to train new and increasingly better models being released at an increasingly fast pace. Here we describe RECODE, a manually annotated corpus of ecological and taxonomic literature, aimed at training and fine-tuning models for automated extraction of occurrence and trait data from unstructured text. All documents presented at this stage have been annotated and validated by experts familiar with the traits of the test taxa (spiders and insects).

Schlagwörter

Stichwörter der Autor:innen

biodiversityfunctional diversityinsectslarge language modelsmachine learningnamed entity recognitionnatural language processingoccurrence dataspecies traitsspiders

Autor:innen und Institutionen

  • Finnish Museum of Natural History LUOMUS, University of Helsinki, Helsinki, Finland Finnish Museum of Natural History LUOMUS, University of Helsinki Helsinki Finland.
  • Centre for Ecology, Evolution and Environmental Changes (cE3c) & CHANGE - Global Change and Sustainability Institute, Faculdade de Ciências, Universidade de Lisboa, Lisbon, Portugal Centre for Ecology, Evolution and Env…
  • LASIGE and Departamento de Informática, Faculdade de Ciências, Universidade de Lisboa, Lisbon, Portugal LASIGE and Departamento de Informática, Faculdade de Ciências, Universidade de Lisboa Lisbon Portugal.
  • University of Helsinki, Helsinki, Finland University of Helsinki Helsinki Finland.
  • Department of Botany and Zoology, Faculty of Science, Masaryk University, Brno, Czech Republic Department of Botany and Zoology, Faculty of Science, Masaryk University Brno Czech Republic.

ORCID-Profile: Vasco V Branco, Lenka Baranovičová, Pedro Cardoso

Zitieren

APA

Branco, V. V., Pivovarova, L., Lintulaakso, K. J., Correia, L. M., Baranovičová, L., Lahin, I., Dias, F., Filipe, D., & Cardoso, P. (2026). RECODE - Relational Ecological COrpus for Data Extraction. Biodiversity Data Journal, 14. https://doi.org/10.3897/bdj.14.e177365

Vancouver

Branco VV, Pivovarova L, Lintulaakso KJ, Correia LM, Baranovičová L, Lahin I, et al. RECODE - Relational Ecological COrpus for Data Extraction. Biodiversity Data Journal. 2026;14. doi:10.3897/bdj.14.e177365

BibTeX
@article{Branco2026,
  author = {Branco, Vasco V and Pivovarova, Lidia and Lintulaakso, Kari-E J and Correia, Luís M and Baranovičová, Lenka and Lahin, Iiris and Dias, Francisco and Filipe, David and Cardoso, Pedro},
  title = {RECODE - Relational Ecological COrpus for Data Extraction},
  journal = {Biodiversity Data Journal},
  year = {2026},
  volume = {14},
  doi = {10.3897/bdj.14.e177365}
}
Kostenlos starten →