Pair-Based Joint Encoding with Relational Graph Convolutional Networks for Emotion-Cause Pair Extraction

TL;DR

This paper introduces PBJE, a joint encoding method using relational graph convolutional networks to improve emotion-cause pair extraction by modeling clause relationships more effectively.

Contribution

The paper proposes a novel joint encoding framework with RGCN for ECPE, addressing imbalance in feature interaction present in prior sequential methods.

Findings

Achieves state-of-the-art performance on Chinese ECPE benchmark.

Effectively models clause and pair relationships with RGCN.

Balances information flow among emotion, cause, and pair features.

Abstract

Emotion-cause pair extraction (ECPE) aims to extract emotion clauses and corresponding cause clauses, which have recently received growing attention. Previous methods sequentially encode features with a specified order. They first encode the emotion and cause features for clause extraction and then combine them for pair extraction. This lead to an imbalance in inter-task feature interaction where features extracted later have no direct contact with the former. To address this issue, we propose a novel Pair-Based Joint Encoding (PBJE) network, which generates pairs and clauses features simultaneously in a joint feature encoding manner to model the causal relationship in clauses. PBJE can balance the information flow among emotion clauses, cause clauses and pairs. From a multi-relational perspective, we construct a heterogeneous undirected graph and apply the Relational Graph Convolutional Network (RGCN) to capture the various relationship between clauses and the relationship between pairs and clauses. Experimental results show that PBJE achieves state-of-the-art performance on the Chinese benchmark corpus.