OffensEval 2020 - Arabic

Organized by hmubarak - Current server time: July 6, 2020, 8:12 a.m. UTC

First phase

Sub-task A
Feb. 19, 2020, midnight UTC

End

Competition Ends
March 5, 2020, 11:59 p.m. UTC

Welcome to OffensEval 2020: Arabic!

Dataset and detailed task description available at: https://sites.google.com/site/offensevalsharedtask/home
 
This is the competition site for the development set of the second edition of OffensEval organized at SemEval 2020 (Task 12). SemEval 2020 will take place on September 13 and 13, 2020 co-located with COLING in Barcelona, Spain. 
 
This year OffensEval's title is Multilingual Offensive Language Identification in Social Media.
 
The first OffensEval was organized at SemEval 2019OffensEval 2019 used the Offensive Language Identification dataset (OLID) a dataset containing English tweets annotated using a hierarchical three-level annotation model described in this paperNearly 800 teams signed up to participate in OffensEval 2019. The competition received more than 100 submissions across three sub-tasks.The findings are described in the OffensEval 2019 reportThe response received in 2019 by far exceeded our expectations and motivated us to organize OffensEval 2020. 

Evaluation Criteria

Submissions will be evaluated for macro F1-score under subtask A. The labels are binary, OFF or NOT, for offensive and not offensive. The classes are imbalanced, hence use of F1.

Terms and Conditions

The data in this competition, OffensEval 2020 - Arabic, is licensed under a Creative Commons Attribution license (CC-BY).

Sub-task A

Start: Feb. 19, 2020, midnight

Competition Ends

March 5, 2020, 11:59 p.m.

You must be logged in to participate in competitions.

Sign In