Natural Language Processing (NLP) is at the core of most information processing tasks. It has a long history in computer science and it has always been a central discipline of Artificial Intelligence. Recently, with the availability of data and the development of processing technologies (such as GPUs, TPUs, ...), we witness a great increasing success of machine and deep learning algorithms, which has marked the renewal of AI and opened its new era.
The success of deep learning algorithms in general and transformers in particular has greatly boosted NLP tasks and changed the landscape of the field that has never been fruitful as today, both in the monolingual and the multilingual dimension.
In this context, CERIST proposes this NLP challenge to students and teams to solve some hot topic specific tasks. These tasks deal with scientific information and social networks information. We'll provide training and test datasets and we expect from teams to propose deep learning approaches to solve the tasks. A form will be available online for registration. Tasks are described bellow:
Task1: Opinion mining and Sentiment Analysis.
task1.a. Arabic Sentiment Analysis and Opinion Mining in book reviews.
task1.b. Multilingual Sentiment Analysis in twitter (with Arabic).
task1.c. Arabic sentiment analysis and fake news detection within covid-19.
task.d. Arabic hate speech and offensive language detection on social networks.
Task2: Information Retrieval
task2.a. Scientific paper classification.
- English (monolingual).
- Arabic (monolingual).
task2.b. Covid-19 literature classification.
task2.c. Automatic text summarization (Arabic and English)
task2.d. Covid- 19 Information Retrieval.
Dates:
- first call: May 23th, 2022.
- Deadline for teams registration: June 30th, 2022.
- Data availability: june 15th, 2022.
- Second call: june 15th , 2022.
- Deadline for project submissions: September 30th, 2022.
- Deadline for paper submission: October 15th, 2022.
- Challenge days ( December 15-17, 2022).
Committees: TBA
If interested, please contact us on: This email address is being protected from spambots. You need JavaScript enabled to view it.