LINGUIST List 32.1537
Mon May 03 2021
Confs: Comp Ling/Online
Editor for this issue: Lauren Perkins <laurenlinguistlist.org>
Martin Krallinger <krallinger.martin
Medical Documents Profession Recognition shared task at IberLEF/SEPLN2021 E-mail this message to a friend
Medical Documents Profession Recognition shared task at IberLEF/SEPLN2021
Short Title: MEDDOPROF
Date: 22-Sep-2021 - 22-Sep-2021
Location: Virtual, Spain
Contact: Martin Krallinger
Contact Email: < click here to access email >
Meeting URL: https://temu.bsc.es/meddoprof/
Linguistic Field(s): Computational Linguistics
We are organizing the first shared task focusing on automatic recognition of professions and occupational status (and normalization to standard multilingual terminologies) in medical documents.
The relevance of text mining of professions and occupational status encompasses multiple human-interest areas, from health and social services, competitive intelligence, human resources, legal NLP and even gender studies.
The need to implement advanced NER systems to detect professions in medical texts has been underscored by the current pandemic, in which the risk of selected occupational groups has resulted in higher mortality and morbidity for these segments of the population. The relationships between disorders and professions may be explained by different factors like increased contact/exposure to hazardous substances, allergens or pathogens; physical injuries due to occupational accidents; higher degrees of social interaction of some professions, or even work-related conditions affecting mental health, just to name a few.
Additionally, targeted vaccination plans also benefit from better characterization of patient professions.
Following the success of previously organized shared tasks (i.e. Cantemist, PharmaCoNER, or Meddocan), we are now launching the MEDDOPROF shared task as part of the IberLEF 2021 evaluation initiative (co-located with SEPLN 2021), with the following sub-tracks:
MEDDOPROF-NER: automatic detection of mentions of occupations (profession, employment status and activities).
MEDDOPROF-CLASS: finding mentions of occupations and classifying them, whether they refer to the patients themselves, their family members or healthcare professionals.
MEDDOPROF-NORM: mapping detected occupation mentions to their corresponding concept identifiers from standard multilingual occupation terminologies (ESCO and SNOMED-CT).
MEDDOPROF web: https://temu.bsc.es/meddoprof/
Annotation guidelines: https://doi.org/10.5281/zenodo.4694675
Google Group for updates: https://groups.google.com/g/meddoprof-shared-task
Test set release (start of evaluation period): June 1, 2021
End of evaluation period (system submissions): June 7, 2021
Working papers submission: June 21, 2021
Notification of acceptance (peer-reviews): June 27, 2021
Camera-ready system descriptions: July 4, 2021
SEPLN 2021: September 2021
Publications and IBERLEF/SEPLN2021 workshop
Teams participating in MEDDOPROF will be invited to contribute a systems description paper for the IberLEF (SEPLN 2021) Working Notes proceedings, and a short presentation of their approach at the IberLEF 2021 workshop.
Martin Krallinger, Barcelona Supercomputing Center, Spain
Eulàlia Farré, Barcelona Supercomputing Center, Spain
Salvador Lima, Barcelona Supercomputing Center, Spain
Vicent Briva-Iglesias, D-REAL, Dublin City University, Ireland
Antonio Miranda-Escalada, Barcelona Supercomputing Center, Spain
Page Updated: 03-May-2021