Urban Incident Prediction with Graph Neural Networks: Integrating Government Ratings and Crowdsourced Reports

TL;DR

This paper introduces a multiview GNN model that combines sparse government ratings and dense crowdsourced reports to accurately predict urban incident states, addressing bias and data sparsity.

Contribution

The paper proposes a novel multiview, multioutput GNN model integrating biased crowdsourced data and sparse official ratings for urban incident prediction.

Findings

Model outperforms single-data-source models in predicting latent incident states.

Crowdsourced reports are biased by neighborhood income levels.

The dataset includes over 9.6 million reports and 1 million ratings for NYC.

Abstract

Graph neural networks (GNNs) are widely used in urban spatiotemporal forecasting, such as predicting infrastructure problems. In this setting, government officials wish to know in which neighborhoods incidents like potholes or rodent issues occur. The true state of incidents (e.g., street conditions) for each neighborhood is observed via government inspection ratings. However, these ratings are only conducted for a sparse set of neighborhoods and incident types. We also observe the state of incidents via crowdsourced reports, which are more densely observed but may be biased due to heterogeneous reporting behavior. First, for such settings, we propose a multiview, multioutput GNN-based model that uses both unbiased rating data and biased reporting data to predict the true latent state of incidents. Second, we investigate a case study of New York City urban incidents and collect, standardize, and make publicly available a dataset of 9,615,863 crowdsourced reports and 1,041,415 government inspection ratings over 3 years and across 139 types of incidents. Finally, we show on both real and semi-synthetic data that our model can better predict the latent state compared to models that use only reporting data or models that use only rating data, especially when rating data is sparse and reports are predictive of ratings. We also quantify demographic biases in crowdsourced reporting, e.g., higher-income neighborhoods report problems at higher rates. Our analysis showcases a widely applicable approach for latent state prediction using heterogeneous, sparse, and biased data.