EmoContext: SEMEVAL’19
We want to analyse the sentiment of an utterance of a user, eg:- a tweet, a reply, etc., leveraging the lexical, semantic and contextual information of the preceding conversation. Also, we’d like to know what caused that sentiment, some trigger words, the sentiment of a previous utterance, etc. This is an ongoing challenge for SEMEVAL’19.
“Understanding Emotions in Textual Conversations is a hard problem in absence of voice modulations and facial expressions.” Most of the solutions analysing sentiment treat the utterances as individual entities w/o leveraging the context of the whole conversation, which can completely change the meaning of an utterance. Also in this age of micro-blogging, the strict bound on word limit makes it a much more challenging task.
Given a textual dialogue i.e. a user utterance along with previous two utterances of the conversation, the task is to classify the emotion of the final user utterance as one of the emotion classes: Happy, Sad, Angry or Others. Also, learn the class w/o the previous conversation (For PLT ).
We’re using the dataset provided in the challenge.