Welcome to EmotionX 2019, the shared task of SocialNLP 2019.
Participants are challenged to predict the emotions induced by dialogue utterances by classifying each utterance into one of four label candidates: joy, sadness, anger, and neutral. We include both spoken and chat-based dialogues. Read more about the task.
NEW! For 2019, we used Google Translate to augment our data using double-translated utterances.
Participants must register in order to participate.
Note: Technical Papers for this task will not be archived
Note: Participants are gently encouraged to share their code on GitHub or similar platform.