SEOVER: Sentence-level Emotion Orientation Vector based Conversation Emotion Recognition Model

TL;DR

This paper introduces SEOVER, a novel sentence-level emotion orientation vector approach for conversation emotion recognition, which improves accuracy by modeling emotional tendencies and correlations between utterances.

Contribution

It proposes a new emotion representation paradigm and a joint learning model that enhances emotion recognition performance in conversations.

Findings

Outperforms five baseline models on benchmark datasets.

Improves emotion recognition accuracy across all tested datasets.

Demonstrates the effectiveness of modeling emotional tendencies at sentence level.

Abstract

For the task of conversation emotion recognition, recent works focus on speaker relationship modeling but ignore the role of utterance's emotional tendency.In this paper, we propose a new expression paradigm of sentence-level emotion orientation vector to model the potential correlation of emotions between sentence vectors. Based on it, we design an emotion recognition model, which extracts the sentence-level emotion orientation vectors from the language model and jointly learns from the dialogue sentiment analysis model and extracted sentence-level emotion orientation vectors to identify the speaker's emotional orientation during the conversation. We conduct experiments on two benchmark datasets and compare them with the five baseline models.The experimental results show that our model has better performance on all data sets.