CrimeForecaster: Crime Prediction by Exploiting the Geographical Neighborhoods’ Spatiotemporal Dependencies

Sun, Jiao; Yue, Mingxuan; Lin, Zongyu; Yang, Xiaochen; Nocera, Luciano; Kahn, Gabriel; Shahabi, Cyrus

doi:10.1007/978-3-030-67670-4_4

Cited by 14 publications

(19 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…In this paper, we focus on the task of crime forecasting given previous crime records. Our setting is the same as Crime-Forecaster's (Sun et al 2021).…”

Section: Problem Definitionmentioning

confidence: 99%

“…, y K i,C ) ∈ R C×K to denote all C types of crimes that occurred during the past K slots' observations. Following the general settings of previous crime forecasting works (Huang et al 2018(Huang et al , 2019Sun et al 2021), we set each element y k i,l to 1 if crime type l happens at region i in time slot t, and 0 otherwise.…”

Section: Problem Definitionmentioning

confidence: 99%

“…Data Description and Training Configuration We evaluated HAGEN on two real-world benchmarks in Chicago and Los Angeles by CrimeForecaster (Sun et al 2021). In our experiments, we use the same "train-validation-test" setting as the previous work (Sun et al 2021;Huang et al 2019). We chronologically split the dataset as 6.5 months for training, 0.5 months for the validation, and 1 month for testing.…”

Section: Data and Experimental Settingmentioning

confidence: 99%

“…However, they assume a gridbased partitioning of the underlying geographic region to utilize CNN-based models which ignore the geographical extents of neighborhoods. Based on the intuition that nearby regions (i.e., geographical neighborhoods which collect criminal records within a period of time) have similar crime patterns, Sun et al (2021) leveraged a distance-based region graph to explore the connection between spatial and temporal crime patterns, and utilized a graph neural network (GNN) technique to model the spatial dependencies between regions. In this setting, the crime forecasting problem is equivalent to learning a mapping function to predict the future graph signals (i.e., crime occurrences) given the historical graph signals (i.e., historical crime records) over a distance-based region graph.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

HAGEN: Homophily-Aware Graph Convolutional Recurrent Network for Crime Forecasting

Wang¹,

Lin²,

Yang³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

The crime forecasting is an important problem as it greatly contributes to urban safety. Typically, the goal of the problem is to predict different types of crimes for each geographical region (like a neighborhood or censor tract) in the near future. Since nearby regions usually have similar socioeconomic characteristics which indicate similar crime patterns, recent state-of-the-art solutions constructed a distance-based region graph and utilized Graph Neural Network (GNN) techniques for crime forecasting, because the GNN techniques could effectively exploit the latent relationships between neighboring region nodes in the graph if the edges reveal high dependency or correlation. However, this distance-based predefined graph cannot fully capture crime correlation between regions that are far from each other but share similar crime patterns. Hence, to make an accurate crime prediction, the main challenge is to learn a better graph that reveals the dependencies between regions in crime occurrences and meanwhile captures the temporal patterns from historical crime records. To address these challenges, we propose an endto-end graph convolutional recurrent network called HAGEN with several novel designs for crime prediction. Specifically, our framework could jointly capture the crime correlation between regions and the temporal crime dynamics by combining an adaptive region graph learning module with the Diffusion Convolution Gated Recurrent Unit (DCGRU). Based on the homophily assumption of GNN (i.e., graph convolution works better where neighboring nodes share the same label), we propose a homophily-aware constraint to regularize the optimization of the region graph so that neighboring region nodes on the learned graph share similar crime patterns, thus fitting the mechanism of diffusion convolution. It also incorporates crime embedding to model the interdependencies between regions and crime categories. Empirical experiments and comprehensive analysis on two real-world datasets showcase the effectiveness of HAGEN.Preprint. Under review.

show abstract

“…In this paper, we focus on the task of crime forecasting given previous crime records. Our setting is the same as Crime-Forecaster's (Sun et al 2021).…”

Section: Problem Definitionmentioning

confidence: 99%

Section: Problem Definitionmentioning

confidence: 99%

Section: Data and Experimental Settingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

HAGEN: Homophily-Aware Graph Convolutional Recurrent Network for Crime Forecasting

Wang¹,

Lin²,

Yang³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…Furthermore, they used SRNN like structure with the multi-layered LSTMs on edges and performed predictions for 50 regions in the CHI crime dataset. Another study [21] implemented Gated Recurrent Network with Diffusion Convolution modules following a Multi-Layer Perceptron (MLP). Their experiments on CHI data built the graph according to districts where the edge values are the distance between nodes.…”

Section: A Prior Art and Comparisonsmentioning

confidence: 99%

Crime Prediction with Graph Neural Networks and Multivariate Normal Distributions

Tekin¹,

Kozat²

2021

Preprint

View full text Add to dashboard Cite

Existing approaches to the crime prediction problem are unsuccessful in expressing the details since they assign the probability values to large regions. This paper introduces a new architecture with the graph convolutional networks (GCN) and multivariate Gaussian distributions to perform high-resolution forecasting that applies to any spatiotemporal data. We tackle the sparsity problem in high resolution by leveraging the flexible structure of GCNs and providing a subdivision algorithm. We build our model with Graph Convolutional Gated Recurrent Units (Graph-ConvGRU) to learn spatial, temporal, and categorical relations. In each node of the graph, we learn a multivariate probability distribution from the extracted features of GCNs. We perform experiments on real-life and synthetic datasets, and our model obtains the best validation and the best test score among the baseline models with significant improvements. We show that our model is not only generative but also precise.

show abstract

Spatial-Temporal Attention Network for Crime Prediction with Adaptive Graph Learning

Sun

Zhou

Tian

et al. 2022

Lecture Notes in Computer Science

View full text Add to dashboard Cite

CrimeForecaster: Crime Prediction by Exploiting the Geographical Neighborhoods’ Spatiotemporal Dependencies

Cited by 14 publications

References 15 publications

HAGEN: Homophily-Aware Graph Convolutional Recurrent Network for Crime Forecasting

HAGEN: Homophily-Aware Graph Convolutional Recurrent Network for Crime Forecasting

Crime Prediction with Graph Neural Networks and Multivariate Normal Distributions

Spatial-Temporal Attention Network for Crime Prediction with Adaptive Graph Learning

Contact Info

Product

Resources

About