PESE: Event Structure Extraction using Pointer Network based Encoder-Decoder Architecture

TL;DR

This paper introduces PESE, a pointer network-based encoder-decoder model that extracts comprehensive event structures from text in an end-to-end manner, capturing interdependencies among event components.

Contribution

The paper proposes a novel end-to-end approach using pointer networks to extract complete event tuples, improving over methods that treat substructures separately.

Findings

Achieves competitive performance on ACE2005 dataset

Effectively models interdependencies among event components

Outperforms some existing state-of-the-art methods

Abstract

The task of event extraction (EE) aims to find the events and event-related argument information from the text and represent them in a structured format. Most previous works try to solve the problem by separately identifying multiple substructures and aggregating them to get the complete event structure. The problem with the methods is that it fails to identify all the interdependencies among the event participants (event-triggers, arguments, and roles). In this paper, we represent each event record in a unique tuple format that contains trigger phrase, trigger type, argument phrase, and corresponding role information. Our proposed pointer network-based encoder-decoder model generates an event tuple in each time step by exploiting the interactions among event participants and presenting a truly end-to-end solution to the EE task. We evaluate our model on the ACE2005 dataset, and experimental results demonstrate the effectiveness of our model by achieving competitive performance compared to the state-of-the-art methods.