Published January 31, 2024 | Version v1
Model Open

Odeuropa information extraction system models version 2

Authors/Creators

  • 1. ROR icon Fondazione Bruno Kessler

Description

This resource contains the BERT based models used by the information extraction system within the Odeuropa European project (https://odeuropa.eu/)

These model can be used to extract from texts the presence of smell related words, smell sources , qualities, perceivers, circumstances, odor carriers, effects provoked by the smell, evoked odorants, places and time.

We provide a different model for each language covered by the Odeuropa project: English, Italian, French, German, Dutch and Slovene

The code of the smell information extraction system to use the models is available in the repository:

https://github.com/Odeuropa/wp3-information-extraction-system-v2

 

Files

de-model-v2.zip

Files (3.3 GB)

Name Size
md5:1fb2c256dab50b78da0beebee1714e10
660.4 MB Preview Download
md5:99eba3ddb803fce1df7e67d80c61307a
402.3 MB Preview Download
md5:c33880a2a5cec7127d910e85ae314691
513.6 MB Preview Download
md5:15ad0670b65cf0e661aa32820d4638ff
660.4 MB Preview Download
md5:9d03c0f3ba8b8ab69c8ea522abbf438c
660.4 MB Preview Download
md5:a1827c1b7cb7016b02c15f7092dbe518
402.1 MB Preview Download

Additional details

Funding

European Commission
ODEUROPA - ODEUROPA: Negotiating Olfactory and Sensory Experiences in Cultural Heritage Practice and Research 101004469

References

  • Menini, Stefano. Semantic Frame Extraction in Multilingual Olfactory Events. In Proceedings of LREC-Coling 2024