Semi-Supervised Graph Attention Networks for Event Representation Learning

TL;DR

This paper introduces GNEE, a semi-supervised graph attention network method for event representation learning that effectively handles heterogeneous event graphs and leverages labeled data for improved embeddings.

Contribution

The paper proposes GNEE, combining Graph Attention Networks and graph regularization to address heterogeneity and semi-supervised learning in event graph embeddings.

Findings

GNEE outperforms existing semi-supervised graph embedding methods.

The method effectively handles graph heterogeneity.

Self-attention improves relationship importance learning.

Abstract

Event analysis from news and social networks is very useful for a wide range of social studies and real-world applications. Recently, event graphs have been explored to model event datasets and their complex relationships, where events are vertices connected to other vertices representing locations, people's names, dates, and various other event metadata. Graph representation learning methods are promising for extracting latent features from event graphs to enable the use of different classification algorithms. However, existing methods fail to meet essential requirements for event graphs, such as (i) dealing with semi-supervised graph embedding to take advantage of some labeled events, (ii) automatically determining the importance of the relationships between event vertices and their metadata vertices, as well as (iii) dealing with the graph heterogeneity. This paper presents GNEE (GAT Neural Event Embeddings), a method that combines Graph Attention Networks and Graph Regularization. First, an event graph regularization is proposed to ensure that all graph vertices receive event features, thereby mitigating the graph heterogeneity drawback. Second, semi-supervised graph embedding with self-attention mechanism considers existing labeled events, as well as learns the importance of relationships in the event graph during the representation learning process. A statistical analysis of experimental results with five real-world event graphs and six graph embedding methods shows that our GNEE outperforms state-of-the-art semi-supervised graph embedding methods.