Label Enhanced Event Detection with Heterogeneous Graph Attention Networks

TL;DR

This paper introduces L-HGAT, a novel graph-based model that enhances Chinese event detection by fully exploiting character-word interactions and incorporating event label semantics, leading to improved performance.

Contribution

The paper proposes a new heterogeneous graph attention network that models character-word interactions and integrates label semantics for better Chinese event detection.

Findings

Significant performance improvements over baseline methods.

Effective modeling of character-word interactions.

Utilization of label semantics enhances detection accuracy.

Abstract

Event Detection (ED) aims to recognize instances of specified types of event triggers in text. Different from English ED, Chinese ED suffers from the problem of word-trigger mismatch due to the uncertain word boundaries. Existing approaches injecting word information into character-level models have achieved promising progress to alleviate this problem, but they are limited by two issues. First, the interaction between characters and lexicon words is not fully exploited. Second, they ignore the semantic information provided by event labels. We thus propose a novel architecture named Label enhanced Heterogeneous Graph Attention Networks (L-HGAT). Specifically, we transform each sentence into a graph, where character nodes and word nodes are connected with different types of edges, so that the interaction between words and characters is fully reserved. A heterogeneous graph attention networks is then introduced to propagate relational message and enrich information interaction. Furthermore, we convert each label into a trigger-prototype-based embedding, and design a margin loss to guide the model distinguish confusing event labels. Experiments on two benchmark datasets show that our model achieves significant improvement over a range of competitive baseline methods.