BioASQ Participants Area
BioASQ - Task BioNNE
The proposed BioNNE shared tasks involve NLP challenges on biomedical nested named entity recognition (NER) systems for English and Russian languages. The train/dev datasets include annotated mentions of disorders, anatomical structures, chemicals, diagnostic procedures, and biological functions. Participants are allowed to train any model architecture on any publicly available data in order to achieve the best performance.
The evaluation framework is divided into three broad tracks:
Track 1 - Bilingual: Participants in this track are required to train a single multi-lingual NER model using training data for both Russian and English languages. The model should be used to generate prediction files for each language's dataset.
Track 2 - English-oriented: Participants in this track are required to train a nested NER model for English scientific abstracts in the biomedical domain.
Track 3 - Russian-oriented: Participants in this track are required to train a nested NER model for Russian scientific abstracts in the biomedical domain.
See the details in our GitHub.
To submit your results you need to register on the Codalab Competition page.
The training data for the BioNNE task are available here.
Timeline:| Phase | Dates |
| Registration opens | 13 Nov 2023 |
| Training Data Release | 5 Feb 2024 |
| Test data release, evaluation phase starts | 25 April 2024 |
| Test set predictions due | 6 May 2024 |
| Submission of participant papers | 31 May 2024 |
| Acceptance notification for participant papers | 24 June 2024 |
| Camera-ready working notes papers | 8 July 2024 |
| BioASQ Workshop at CLEF 2024, Grenoble, France | September 9-12, 2024 |
References:
Loukachevitch, N., Manandhar, S., Baral, E., Rozhkov, I., Braslavski, P., Ivanov, V., ... & Tutubalina, E. (2023). NEREL-BIO: a dataset of biomedical abstracts annotated with nested named entities.
Bioinformatics, 39(4), btad161. https://doi.org/10.1093/bioinformatics/btad161
