ERNetCL: A novel emotion recognition network in textual conversation based on curriculum learning strategy

TL;DR

This paper introduces ERNetCL, a new emotion recognition network in textual conversations that uses curriculum learning to better capture context and improve performance over existing models.

Contribution

It proposes a novel ERC model combining temporal and spatial encoders with curriculum learning, enhancing context modeling and training efficiency.

Findings

ERNetCL outperforms baseline models on four datasets.

The curriculum learning strategy improves training stability and accuracy.

The model effectively captures temporal and spatial contextual cues.

Abstract

Emotion recognition in conversation (ERC) has emerged as a research hotspot in domains such as conversational robots and question-answer systems. How to efficiently and adequately retrieve contextual emotional cues has been one of the key challenges in the ERC task. Existing efforts do not fully model the context and employ complex network structures, resulting in limited performance gains. In this paper, we propose a novel emotion recognition network based on curriculum learning strategy (ERNetCL). The proposed ERNetCL primarily consists of temporal encoder (TE), spatial encoder (SE), and curriculum learning (CL) loss. We utilize TE and SE to combine the strengths of previous methods in a simplistic manner to efficiently capture temporal and spatial contextual information in the conversation. To ease the harmful influence resulting from emotion shift and simulate the way humans learn curriculum from easy to hard, we apply the idea of CL to the ERC task to progressively optimize the network parameters. At the beginning of training, we assign lower learning weights to difficult samples. As the epoch increases, the learning weights for these samples are gradually raised. Extensive experiments on four datasets exhibit that our proposed method is effective and dramatically beats other baseline models.