WojoodNER 2024 The 2nd Arabic Named Entity Recognition Shared Task at ArabicNLP’24
ندعوكم للمشاركة في المسابقة العلمية الثانية لاكتشاف الاعلام في النصوص العربية. سيحصل المشاركين على مدونة وجود الجديدة (٥٥٠ الف كلمة + انواع مفصلة من الاعلام). يوجد ثلاث مهام في المسابقة يمكن المشاركة باي منها، احدى المهام حول الحرب على غزة ويمكن للمشاركين استخدام بيانات خارجية فيها
Dataset: Wojood-Fine https://aclanthology.org/2023.arabicnlp-1.25/ New version: Arabic Fine-Grained Entity Recognition (Wojood + Subtypes of entity types).
Subtask-1 (Closed-Track Flat Fine-Grain NER): We provide the Wojood-Fine Flat train (70%) and development (10%) datasets. The final evaluation will be on the test set (20%). External data is not allowed .... (read more https://dlnlp.ai/st/wojood/).
Subtask-2 (Closed-Track Nested Fine-Grain NER): This subtask is similar to the subtask-1, we provide the Wojood-Fine Nested train (70%) and development (10%) datasets. The final evaluation will be on the test set (20%) .... (read more https://dlnlp.ai/st/wojood/).
Subtask-3 (Open-Track NER - Gaza War): to allow participants to reflect on the utility of NER in the context of real-world events, allow them to use external resources, and encourage them to use generative models in different ways (fine-tuned, zero-shot learning, in-context learning, etc.). The goal of focusing on generative models in this particular subtask is to help the Arabic NLP research community better understand the capabilities and performance gaps of LLMs in information extraction, an area currently understudied. We provide development and test data related to the current War on Gaza. This is motivated by the assumption that discourse about recent global events will involve mentions from different data distribution. For this subtask, we include data from five different news domains related to the War on Gaza - but we keep the names of the domains hidden. Participants will be given a development dataset (10K tokens, 2K from each of the five domains), and a testing dataset (50K tokens, 10K from each domain). Both development and testing sets are manually annotated with fine-grain named entities using the same annotation guidelines used in Subtask1 and Subtask2 (also described in Liqreina et al., 2023). .... (read more https://dlnlp.ai/st/wojood/).
BASELINES
Two baseline models trained on WojoodFine (flat and nested) are provided (See Liqreina et al., 2023 https://aclanthology.org/2023.arabicnlp-1.25/). The code used to produce these baselines is available on GitHub https://github.com/SinaLab/ArabicNER.
Subtask Precision Recall Average Micro-F1 Flat Fine-Grain NER (Subtask 1) 0.8870 0.8966 0.8917 Nested Fine-Grain NER (Subtask 2) 0.9179 0.9279 0.9229 GOOGLE COLAB NOTEBOOKS
To allow you to experiment with the baseline, we authored four Google Colab notebooks that demonstrate how to train and evaluate our baseline models. [1] Train Flat Fine-Grain NER https://gist.github.com/mohammedkhalilia/72c3261734d7715094089bdf4de74b4a: This notebook can be used to train our ArabicNER model on the flat Fine-grain NER task using the sample Wojood_Fine data. [2] Evaluate Flat Fine-Grain NER https://gist.github.com/mohammedkhalilia/c807eb1ccb15416b187c32a362001665: This notebook will use the trained model saved from the notebook above to perform evaluation on unseen dataset. [3] Train Nested Fine-Grain NER https://gist.github.com/mohammedkhalilia/a4d83d4e43682d1efcdf299d41beb3da: This notebook can be used to train our ArabicNER model on the nested Fine-grain task using the sample Wojood data. [4] Evaluate Nested Fine-Grain NER https://gist.github.com/mohammedkhalilia/9134510aa2684464f57de7934c97138b: This notebook will use the trained model saved from the notebook above to perform evaluation on unseen dataset.
REGISTRATION
Participants need to register via this form (NERSharedTask 2024) https://docs.google.com/forms/d/1ISMILgQYfUug3XuDpxFmuPASXkWaduYOUc3xOZuGwqU/edit?ts=65a82a3a. Participating teams will be provided with common training development datasets. No external manually labelled datasets are allowed. Blind test data set will be used to evaluate the output of the participating teams. Each team is allowed a maximum of 3 submissions. All teams are required to report on the development and test sets (after results are announced) in their write-ups.
FAQ
For any questions related to this task, please check our Frequently Asked Questions https://docs.google.com/document/d/1W_13FRpP3NbDx_ALYJWA3-ESXPRVomOjNovUuYfdmI0/edit?usp=sharing IMPORTANT DATES
- February 25, 2024: Shared task announcement. - March 1, 2024: Release of training data, development sets, scoring script, and Codalab links. - April 5, 2024: Registration deadline. - April 26, 2024: Test set made available. - May 3, 2024: Codalab Test system submission deadline. - May 10, 2024: Shared task system paper submissions due. - June 17, 2024: Notification of acceptance. - July 1, 2024: Camera-ready version. - August 16, 2024: ArabicNLP 2024 conference in Thailand.
CONTACT
For any questions related to this task, please contact the organizers directly using the following email address: NERSharedtask@gmail.com mailto:NERSharedtask@gmail.com .
ORGANIZERS
- Mustafa Jarrar, Birzeit University - Muhammad Abdul-Mageed, University of British Columbia & MBZUAI - Mohammed Khalilia, Birzeit University - Bashar Talafha, University of British Columbia - AbdelRahim Elmadany, University of British Columbia - Nagham Hamad, Birzeit University
--Mustafa __________________________ Mustafa Jarrar, PhD Professor of Artificial Intelligence Chair, PhD Program in Computer Science Birzeit University, Palestine Whatsapp:+972599662258 | mjarrar@birzeit.edu mailto:mjarrar@birzeit.edu http://www.jarrar.info http://www.jarrar.info/