TempoKGAT: A Novel Graph Attention Network Approach for Temporal Graph Analysis

TL;DR

TempoKGAT is a new graph attention network that effectively models temporal graph data by incorporating time decay and neighbor selection, leading to improved prediction accuracy across various domains.

Contribution

Introduces TempoKGAT, a novel GNN that combines time-decaying weights and neighbor selection for better temporal graph analysis.

Findings

Outperforms state-of-the-art methods on multiple datasets

Achieves superior accuracy in traffic, energy, and health data

Provides enhanced interpretability of temporal graph models

Abstract

Graph neural networks (GNN) have shown significant capabilities in handling structured data, yet their application to dynamic, temporal data remains limited. This paper presents a new type of graph attention network, called TempoKGAT, which combines time-decaying weight and a selective neighbor aggregation mechanism on the spatial domain, which helps uncover latent patterns in the graph data. In this approach, a top-k neighbor selection based on the edge weights is introduced to represent the evolving features of the graph data. We evaluated the performance of our TempoKGAT on multiple datasets from the traffic, energy, and health sectors involving spatio-temporal data. We compared the performance of our approach to several state-of-the-art methods found in the literature on several open-source datasets. Our method shows superior accuracy on all datasets. These results indicate that TempoKGAT builds on existing methodologies to optimize prediction accuracy and provide new insights into model interpretation in temporal contexts.