Salient event detection via hypergraph convolutional network with cross-view self-supervised learning

• Published in: Neurocomputing (Amsterdam) Vol. 612; p. 128639
• Main Authors: Zhu, Enchang, Yu, Zhengtao, Huang, Yuxin, Gao, Shengxiang, Xian, Yantuan
• Format: Journal Article
• Language: English
• Published: Elsevier B.V 07.01.2025
• Subjects: Event representation; Graph neural network; Hypergraph learning; Salient event detection; Self-supervised learning; Salient event detection; Hypergraph learning; Event representation; Graph neural network; Self-supervised learning
• ISSN: 0925-2312
• DOI: 10.1016/j.neucom.2024.128639

Identifying the most dominant and central events in a document is critical for holistically understanding its important information. To measure the importance of an event, it is critical to understand its context: who is involved, where it happened, which other events it is related to, and what kind of relationship it has, among other factors. Although existing studies have achieved some accomplishments, they are still not fully effective for two main reasons: (1) They incorporate only the discrete global features of the document, which is insufficient for effectively capturing the contextual information of events; (2) They inadequately model the dependency relationships between events. According to previous research findings, it has been shown that hypergraphs effectively capture the global context (i.e., the document-level context) of long text. However, their potential for salient event detection has remained unexplored. To address this, we propose a novel framework called Salient Event Detection via Hypergraph Convolutional Network with Graph Self-supervised Learning (SEDGS). More specifically, we first construct two hypergraphs: one in the event argument view and another in the event view. We then propose a hypergraph convolutional network to model the event context and discourse relations between events. Moreover, to enhance hypergraph modeling and ensure consistency between argument-view and event-view event representations, we employ contrastive self-supervised learning (SSL) in our model training. Experimental results on a standard event salience dataset verify the superiority of SEDGS, advancing state-of-the-art models. •A Hypergraph Learning Model based on Cross-view Self-supervised Learning for Salient Event Detection (SEDGS) is proposed.•We design argument-view hypergraphs and event-view hypergraphs to model event context information and inter-event correlation information, respectively.•We design a novel cross-view self-supervised learning paradigm in model training to enhance hypergraph modeling and improve the salient event detection task.•Experimental results on the publicly available SED dataset demonstrate the superiority of our proposed method.