Deep Learning for Crime Forecasting: The Role of Mobility at Fine-grained Spatiotemporal Scales

2025-09-25

AI Summary

This study addresses short-term crime prediction at fine-grained spatiotemporal scales. Methodologically, we propose a deep learning framework that jointly models micro-level human mobility dynamics, historical crime incidents, and sociodemographic attributes. We design a ConvLSTM architecture to capture spatiotemporal dependencies across heterogeneous data sources and systematically benchmark against logistic regression, random forest, and standard LSTM baselines. Our key contribution is the first systematic integration of high-resolution human mobility data into short-sequence crime forecasting, revealing distinct enhancement mechanisms—human mobility improves violent crime prediction primarily through temporal proximity and property crime prediction via spatial concentration patterns. Experiments on data from four U.S. cities demonstrate statistically significant improvements in recall, precision, and F1-score over state-of-the-art baselines, overcoming fundamental performance bottlenecks of conventional models. The framework advances real-time crime prediction toward higher accuracy and greater interpretability.

Technology Category

• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Data Mining & Knowledge Management: Mining of Spatial, Temporal or Spatio-Temporal Data

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs

Abstract

Objectives: To develop a deep learning framework to evaluate if and how incorporating micro-level mobility features, alongside historical crime and sociodemographic data, enhances predictive performance in crime forecasting at fine-grained spatial and temporal resolutions. Methods: We advance the literature on computational methods and crime forecasting by focusing on four U.S. cities (i.e., Baltimore, Chicago, Los Angeles, and Philadelphia). We employ crime incident data obtained from each city's police department, combined with sociodemographic data from the American Community Survey and human mobility data from Advan, collected from 2019 to 2023. This data is aggregated into grids with equally sized cells of 0.077 sq. miles (0.2 sq. kms) and used to train our deep learning forecasting model, a Convolutional Long Short-Term Memory (ConvLSTM) network, which predicts crime occurrences 12 hours ahead using 14-day and 2-day input sequences. We also compare its performance against three baseline models: logistic regression, random forest, and standard LSTM. Results: Incorporating mobility features improves predictive performance, especially when using shorter input sequences. Noteworthy, however, the best results are obtained when both mobility and sociodemographic features are used together, with our deep learning model achieving the highest recall, precision, and F1 score in all four cities, outperforming alternative methods. With this configuration, longer input sequences enhance predictions for violent crimes, while shorter sequences are more effective for property crimes. Conclusion: These findings underscore the importance of integrating diverse data sources for spatiotemporal crime forecasting, mobility included. They also highlight the advantages (and limits) of deep learning when dealing with fine-grained spatial and temporal scales.

Problem

• Evaluating how mobility data improves crime forecasting accuracy
• Developing deep learning models for fine-grained spatiotemporal crime prediction
• Comparing model performance across different crime types and cities

Innovation

• ConvLSTM network for spatiotemporal crime prediction
• Integrates mobility, crime, and sociodemographic data features
• Uses fine-grained grid data for 12-hour crime forecasting