EmotionX-IDEA: Emotion BERT -- an Affectional Model for Conversation

TL;DR

This paper adapts BERT for dialogue emotion recognition by transforming conversations into causal utterance pairs, achieving high F1 scores on benchmark datasets.

Contribution

It introduces a novel approach to adapt BERT for continuous dialogue emotion prediction using causal utterance pairs.

Findings

Achieved 0.815 micro F1 on Friends dataset.

Achieved 0.885 micro F1 on EmotionPush dataset.

Demonstrated improved emotion recognition accuracy.

Abstract

In this paper, we investigate the emotion recognition ability of the pre-training language model, namely BERT. By the nature of the framework of BERT, a two-sentence structure, we adapt BERT to continues dialogue emotion prediction tasks, which rely heavily on the sentence-level context-aware understanding. The experiments show that by mapping the continues dialogue into a causal utterance pair, which is constructed by the utterance and the reply utterance, models can better capture the emotions of the reply utterance. The present method has achieved 0.815 and 0.885 micro F1 score in the testing dataset of Friends and EmotionPush, respectively.