GSA-Forecaster: Forecasting Graph-Based Time-Dependent Data with Graph Sequence Attention

TL;DR

GSA-Forecaster is a novel deep learning model that employs graph sequence attention to effectively forecast graph-based, time-dependent data by capturing spatial, temporal, and auxiliary information, outperforming existing models.

Contribution

The paper introduces GSA-Forecaster, a new model with a novel graph sequence attention mechanism that better captures temporal dependencies in graph-based forecasting tasks.

Findings

GSA-Forecaster outperforms existing models on real-world datasets.

The graph sequence attention mechanism improves temporal dependency modeling.

Incorporating auxiliary information enhances prediction accuracy.

Abstract

Forecasting graph-based, time-dependent data has broad practical applications but presents challenges. Effective models must capture both spatial and temporal dependencies in the data, while also incorporating auxiliary information to enhance prediction accuracy. In this paper, we identify limitations in current state-of-the-art models regarding temporal dependency handling. To overcome this, we introduce GSA-Forecaster, a new deep learning model designed for forecasting in graph-based, time-dependent contexts. GSA-Forecaster utilizes graph sequence attention, a new attention mechanism proposed in this paper, to effectively manage temporal dependencies. GSA-Forecaster integrates the data's graph structure directly into its architecture, addressing spatial dependencies. Additionally, it incorporates auxiliary information to refine its predictions further. We validate its performance using real-world graph-based, time-dependent datasets, where it demonstrates superior effectiveness compared to existing state-of-the-art models.