"Killing Me" Is Not a Spoiler: Spoiler Detection Model using Graph Neural Networks with Dependency Relation-Aware Attention Mechanism

TL;DR

This paper introduces SDGNN, a syntax-aware graph neural network model that effectively utilizes dependency relations for spoiler detection, outperforming existing models on benchmark datasets.

Contribution

The paper presents a novel spoiler detection model using dependency relation-aware attention in graph neural networks, addressing limitations of previous attention-based models.

Findings

SDGNN outperforms existing spoiler detection models on benchmark datasets.

Dependency relations are crucial for improving spoiler detection accuracy.

Syntax-aware GNNs enhance the utilization of linguistic structures in NLP tasks.

Abstract

Several machine learning-based spoiler detection models have been proposed recently to protect users from spoilers on review websites. Although dependency relations between context words are important for detecting spoilers, current attention-based spoiler detection models are insufficient for utilizing dependency relations. To address this problem, we propose a new spoiler detection model called SDGNN that is based on syntax-aware graph neural networks. In the experiments on two real-world benchmark datasets, we show that our SDGNN outperforms the existing spoiler detection models.