Description, Registration and access to the resources of the task are now available at https://competitions.codalab.org/competitions/19948
Toponym resolution, also known as geoparsing, geo-grounding or place name resolution, aims to assign the correct locations to all names of places mentioned in documents. The automatic toponym resolution in textual documents has multiple applications and, therefore, has been the focus of research for both industrial and academic organizations.
We invite participation in the Toponym Resolution in Scientific Papers challenge as part of SemEval 2019. For this challenge, we chose a scientific domain where the resolution of the names of places is essential: epidemiology. Given the text of a scientific article from PubMed Central, a digital repository archiving biomedical and life sciences journals, you must detect and disambiguate all names of places mentioned. We provide a training set of 105 articles with all toponyms manually annotated and disambiguated by the GeoNames IDs of the expected places the toponyms refer to (GeonNames is a database of geospatial locations and freely available). To get you started, we also provide the code of a strong baseline system detecting toponyms using a feedforward neural network and disambiguating them using a common heuristic, whereby places with the largest population are selected as the correct match.
* Registration (now - January 2019) * Register your team at https://competitions.codalab.org/competitions/19948. This task will use the forum provided by codalab to ask questions to us or other participants. If you have specific requests, please contact Davy Weissenbacher <dweissen at pennmedicine.upenn.edu>.
* Practice (now - December 2018) * When you have registered, you will get access to the training/validation data and the Submit/View Results section in Codalab where you will be able to submit and evaluate your results. The main performance measure are slight variations of the standard metrics of Precision, Recall and F-measure.
* Evaluation (January 2019) * You will be able to self-evaluate your software using Codalab.
After the evaluation, participants are required to submit and review task description papers following the SemEval timeline.
Davy Weissenbacher, Ph.D. Research Scientist Department of Biostatistics, Epidemiology and Informatics The Perelman School of Medicine University of Pennsylvania
Level 4 Blockley Hall 423 Guardian Drive Philadelphia, PA 19104
Email: dweissen at pennmedicine.upenn.edu Website (group): https://healthlanguageprocessing.org/ -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: text/html Size: 3747 bytes Desc: not available URL: <https://mailman.uib.no/public/corpora/attachments/20180926/cf5b6f55/attachment.txt>