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

2025/06/10 by Sidhika Balachandar, Balachandar, Sidhika, Shuvom Sadhuka +7
Computer Science · Engineering · Social Sciences · #FOS: Computer and information sciences #Human Mobility and Location-Based Analysis #Machine Learning (cs.LG) #Mobile Crowdsensing and Crowdsourcing #Traffic Prediction and Management Techniques

paper · pdf · doi:10.48550/arxiv.2506.08740

openalex publication_date 2025/06/10 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

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.

Citations

• Bayesian Modeling of Zero-Shot Classifications for Urban Flood Detection
• Principles for Open Data Curation: A Case Study with the New York City 311 Service Request Data
• Redesigning Service Level Agreements: Equity and Efficiency in City Government Operations
• Temporal Graph Benchmark for Machine Learning on Temporal Graphs
• Robust Training of Graph Neural Networks via Noise Governance
• Causally-guided Regularization of Graph Attention Improves Generalizability
• Trucks Don't Mean Trump: Diagnosing Human Error in Image Analysis
• Semi-Supervised Semantic Segmentation Using Unreliable Pseudo-Labels
• Informative Pseudo-Labeling for Graph Neural Networks with Few Labels
• Positional Encoder Graph Neural Networks for Geographic Data
• A Spatial-temporal Graph Deep Learning Model for Urban Flood Nowcasting Leveraging Heterogeneous Community Features
• FlexMatch: Boosting Semi-Supervised Learning with Curriculum Pseudo Labeling
• Heterogeneous Graph Neural Network with Multi-view Representation Learning
• On Positional and Structural Node Features for Graph Neural Networks on Non-attributed Graphs
• NRGNN: Learning a Label Noise-Resistant Graph Neural Network on Sparsely and Noisily Labeled Graphs
• SST-GNN: Simplified Spatio-temporal Traffic forecasting model using Graph Neural Network
• CoMatch: Semi-supervised Learning with Contrastive Graph Regularization
• CASTLE: Regularization via Auxiliary Causal Graph Discovery
• Examining COVID-19 Forecasting using Spatio-Temporal Graph Neural Networks
• Inductive Graph Neural Networks for Spatiotemporal Kriging
• Do Advanced Information Technologies Produce Equitable Government Responses in Coproduction: An Examination of 311 Systems in 15 U.S. Cities
• Improving Graph Attention Networks with Large Margin-based Constraints
• Pseudo-Labeling and Confirmation Bias in Deep Semi-Supervised Learning
• A Cautionary Tail: A Framework and Case Study for Testing Predictive\n Model Validity
• Graph Attention Networks
• Equity in 311 Reporting: Understanding Socio-Spatial Differentials in the Propensity to Complain
• Semi-Supervised Classification with Graph Convolutional Networks
• node2vec: Scalable Feature Learning for Networks
• DeepWalk

Related

• Graph Neural Networks for Road Safety Modeling: Datasets and Evaluations for Accident Analysis
• Machine Learning for Public Good: Predicting Urban Crime Patterns to Enhance Community Safety
• RiskOracle: A Minute-level Citywide Traffic Accident Forecasting Framework
• VisitHGNN: Heterogeneous Graph Neural Networks for Modeling Point-of-Interest Visit Patterns
• Enhancing Graph Neural Networks in Large-scale Traffic Incident Analysis with Concurrency Hypothesis