This is a competition on all-words Word Sense Disambiguation. Data and more information available at http://lcl.uniroma1.it/wsdeval/. The evaluation is carried out on the concatenation of all five datasets ("ALL"). CodaLab evaluation script will automatically split the results by dataset as well. Hence, it is also possible to evaluate on a single dataset if desired, but the gold ids should follow the "ALL" dataset format.
For evaluating your systems, you will have to output your results in a .txt file (formatted as the gold standard of the ALL dataset in the evaluation framework), name it "output.txt" and compress it in a .zip file. Then, go to "Participate -> Submit", fill out information about your system and upload the .zip file. If you want to have your results displayed in the Leaderboard, click on "Submit to Leaderboard" once processed.
Systems will be evaluated using F-Measure (harmonic mean between precision and recall).
More information at http://lcl.uniroma1.it/wsdeval/
Start: Jan. 9, 2016, midnight
You must be logged in to participate in competitions.Sign In