Open Event Extraction from Online Text using a Generative Adversarial Network

TL;DR

This paper introduces AEM, a novel generative adversarial network-based model for extracting structured open-domain events from long texts, outperforming Bayesian models especially on news articles.

Contribution

The paper presents AEM, a new adversarial neural network approach that models events with a Dirichlet prior and uses a generator and discriminator for effective event extraction.

Findings

AEM outperforms baseline models on Twitter and news datasets.

Significant 15% F-measure improvement on news articles.

AEM enables visualization of extracted events through discriminator features.

Abstract

To extract the structured representations of open-domain events, Bayesian graphical models have made some progress. However, these approaches typically assume that all words in a document are generated from a single event. While this may be true for short text such as tweets, such an assumption does not generally hold for long text such as news articles. Moreover, Bayesian graphical models often rely on Gibbs sampling for parameter inference which may take long time to converge. To address these limitations, we propose an event extraction model based on Generative Adversarial Nets, called Adversarial-neural Event Model (AEM). AEM models an event with a Dirichlet prior and uses a generator network to capture the patterns underlying latent events. A discriminator is used to distinguish documents reconstructed from the latent events and the original documents. A byproduct of the discriminator is that the features generated by the learned discriminator network allow the visualization of the extracted events. Our model has been evaluated on two Twitter datasets and a news article dataset. Experimental results show that our model outperforms the baseline approaches on all the datasets, with more significant improvements observed on the news article dataset where an increase of 15\% is observed in F-measure.