Given the text of a tweet, the task consists of predicting the most likely emoji to be used along such tweet. It is a multi-class classification task.
The list of possible classes (emojis) is provided in: https://aclanthology.org/S18-1003.pdf
Publication
Francesco Barbieri, Jose Camacho-Collados, Francesco Ronzano, Luis Espinosa-Anke, Miguel Ballesteros, Valerio Basile, Viviana Patti, Horacio Saggion (2018) SemEval 2018 Task 2: Multilingual Emoji Prediction. Proceedings of the 12th International Workshop on Semantic Evaluation (SemEval-2018), pages 24–33.
Language
Spanish
English
NLP topic
Abstract task
Dataset
Year
2018
Publication link
Ranking metric
Accuracy
Task results
System | Accuracy | MacroPrecision | MacroRecall | MacroF1 Sort ascending |
---|---|---|---|---|
Tübingen-Oslo | 0.3727 | 0.2349 | 0.2280 | 0.2236 |
Hatching Chick | 0.3723 | 0.2066 | 0.1916 | 0.1873 |
UMDuluth-CS8761 | 0.3483 | 0.1902 | 0.1860 | 0.1818 |
TAJJEB | 0.2513 | 0.1899 | 0.2036 | 0.1708 |
Duluth UROP | 0.2851 | 0.1711 | 0.1810 | 0.1675 |