[5 days to go to register!] Call for participation on Propaganda Detection in Arabic shared task at WANLP

[Firoj Alam]

Dear colleagues,

Apologies for cross posting! 
You are invited to submit a system paper for the first Propaganda Detection in Arabic shared task held with the 7th Arabic Natural Language Processing Workshop (WANLP 2022) co-located this with the EMNLP2022 Conference in Abu Dhabi (Dec 7, 2022).
https://sites.google.com/view/propaganda-detection-in-arabic/
 
 
Tasks
Subtask 1: Given only the “text” of a Tweet, identify which of the 20 techniques (https://propaganda.qcri.org/annotations/definitions.html) are used in it. This is a multilabel classification problem.
 
Subtask 2: Given only the “text” of a Tweet, identify which of the 20 techniques (https://propaganda.qcri.org/annotations/definitions.html) are used in it together with the span(s) of text covered by each technique. This is a sequence tagging task
 
Important Dates
July 24, 2022: Release of training, dev and dev-test data, and evaluation scripts.
September 10, 2022: Registration deadline.
September 10, 2022: Release of test data (and final training and dev data).
September 15, 2022: End of the evaluation cycle (test set submission closes)
September 20, 2022: Results released
October 5, 2022: System description paper submissions due.
October 15, 2022: Notification of acceptance.
October 31, 2022: Camera-ready versions due.
December 7, 2022: WANLP 2022 workshop at EMNLP in Abu Dhabi
 
Lab Registration
Please register here: https://docs.google.com/forms/d/e/1FAIpQLScfqL4TUeycPPKgkqqdy_LRHN4vAS2lKiPHU3QYo8rjABWFpA/viewform
 
Datasets
The Datasets are hosted on the Gitlab repository: https://gitlab.com/arabic-nlp/propaganda-detection/
 
Organizers
Firoj Alam, Qatar Computing Research Institute, HBKU
Hamdy Mubarak, Qatar Computing Research Institute, HBKU
Wajdi Zaghouani, HBKU
Preslav Nakov, Qatar Computing Research Institute, HBKU
Giovanni Da San Martino, University of Padova