Graph-Based Deep Modeling and Real Time Forecasting of Sparse Spatio-Temporal Data

TL;DR

This paper introduces a comprehensive framework combining a self-exciting point process and a graph neural network to model and forecast sparse spatio-temporal data, demonstrated on crime and traffic prediction.

Contribution

It presents a novel multi-scale deep learning approach that integrates macro and micro-level modeling for sparse spatio-temporal data forecasting.

Findings

Effective in crime and traffic forecasting

Outperforms existing models in accuracy

Handles data sparsity in space and time

Abstract

We present a generic framework for spatio-temporal (ST) data modeling, analysis, and forecasting, with a special focus on data that is sparse in both space and time. Our multi-scaled framework is a seamless coupling of two major components: a self-exciting point process that models the macroscale statistical behaviors of the ST data and a graph structured recurrent neural network (GSRNN) to discover the microscale patterns of the ST data on the inferred graph. This novel deep neural network (DNN) incorporates the real time interactions of the graph nodes to enable more accurate real time forecasting. The effectiveness of our method is demonstrated on both crime and traffic forecasting.