Dear Colleagues,

We are glad to invite you to participate in SemEval-2023 Shared Task 12, the first SemEval shared task for sentiment analysis targeting African low resource languages. The AfriSenti-SemEval Shared Task 12 is based on a collection of humanly annotated Twitter datasets in 16 African languages (all low resource) for sentiment classification. This year, we have three sub-tasks, from which the participants can choose one or more tasks depending on their preference.

Task A: Monolingual Sentiment Classification

Given training data in a target language, determine the polarity of a tweet in the target language (positive, negative, or neutral). If a tweet conveys both a positive and negative sentiment, whichever is the stronger sentiment should be chosen. This sub-task covers 15 languages namely, Hausa, Yoruba, Igbo, Nigerian_Pidgin, Amharic, Algerian Arabic, Kinyarwanda, Twi, Mozambican Portuguese, Swahili, Setswana, isiZulu, Moroccan Arabic/Darija, Xitsonga (South-African Dialect), Xitsonga (Mozambique Dialect)

Task B: Multilingual Sentiment Classification

Given a combined training data from 12 African languages, determine the polarity of a tweet in the target language (positive, negative, or neutral).

Task C: Zero-Shot Sentiment Classification

Given unlabeled tweets in two African languages (Oromo, and Tigrinya), leverage any or all of the available training datasets in Subtasks A and B to determine the sentiment of a tweet in the two target languages is positive, negative, or neutral.

Helpful Links:

Codalab competition page: https://codalab.lisn.upsaclay.fr/competitions/7320 Task website: https://afrisenti-semeval.github.io/

Prize:

The afriSenti-SemEval competition has a prize and will be awarded to the best performing team in each of the three sub-tasks.

- African League: To encourage African participation, this league is for teams with at least one African .

- Masters and Undergraduate League: This league is dedicated to masters and undergraduate students only .

- Worldwide League: Be a participant from any country.

Important Dates:

- Training data ready - 11 September 2022 - Evaluation Starts - 10 January 2023 - Evaluation End - 31 January 2023 - System Description Paper Due - February 2023 - SemEval workshop Summer 2023 - (co-located with a major NLP conference)

You can reach out to the organizers at afrisenti-semeval-organizers@googlegroups.com

Task organizers:

- Shamsuddeen Hassan Muhammad (University of Porto, Portugal, Bayero University, Kano, MasakhaneNLP) - Seid Muhie Yimam ( Universität Hamburg, Hamburg; MasakhaneNLP) - Idris Abdulmumin ( Ahmadu Bello University Zaria, Masakhane NLP) - Ibrahim Sa’id Ahmad ( Bayero University, Kano) - Abinew Ali Ayele ( Bahir Dar University, Bahir Dar) - David Ifeoluwa Adelani ( Saarland University, MasaKhaneNLP) - Bello Shehu Bello (Bayero University, Kano) - Vukosi Marivate (University of Pretoria; MasaKhane) - Sebastian Ruder (Google Research ) - Saif M. Mohammad (National Research Council,Canada) - Nedjma Ousidhoum ( The University of Cambridge) - Meriem Beloucif (Uppsala University) - Tadesse Destaw Belay (Wollo University, Dessie)