Spatial-Temporal Hypergraph Self-Supervised Learning for Crime Prediction (2022-04-18T00:00:00.000000Z)

TL;DR

This work proposes a Spatial-Temporal Self-Supervised Hypergraph Learning framework (ST-HSL), and designs the dual-stage self-supervised learning paradigm, to not only jointly capture local- and global-level spatial-temporal crime patterns, but also supplement the sparse crime representation by augmenting region self-discrimination.

Abstract

Crime has become a major concern in many cities, which calls for the rising demand for timely predicting citywide crime occurrence. Accurate crime prediction results are vital for the beforehand decision-making of government to alleviate the increasing concern about the public safety. While many efforts have been devoted to proposing various spatial-temporal forecasting techniques to explore dependence across locations and time periods, most of them follow a supervised learning manner, which limits their spatial-temporal representation ability on sparse crime data. Inspired by the recent success in self-supervised learning, this work proposes a Spatial-Temporal Self-Supervised Hypergraph Learning framework (ST-HSL) to tackle the label scarcity issue in crime prediction. Specifically, we propose the cross-region hypergraph structure learning to encode region-wise crime dependency under the entire urban space. Furthermore, we design the dual-stage self-supervised learning paradigm, to not only jointly capture local- and global-level spatial-temporal crime patterns, but also supplement the sparse crime representation by augmenting region self-discrimination. We perform extensive experiments on two real-life crime datasets. Evaluation results show that our ST-HSL significantly outperforms state-of-the-art baselines. Further analysis provides insights into the superiority of our ST-HSL method in the representation of spatial-temporal crime patterns. The implementation code is available at https://github.com/LZH-YS1998/STHSL.

Authors

Lianghao Xia

5 papers

Chao Huang

4 papers

Yong Xu

5 papers

TL;DR

Abstract

Authors

References57 items

Contrastive Meta Learning with Behavior Multiplicity for Recommendation

Self-supervised Representation Learning on Dynamic Graphs

Dynamic and Multi-faceted Spatio-temporal Deep Learning for Traffic Speed Forecasting

Spatial-Temporal Sequential Hypergraph Network for Crime Prediction with Dynamic Multiplex Relation Learning

Recent Advances in Heterogeneous Relation Learning for Recommendation

Self-supervised and Supervised Joint Training for Resource-rich Machine Translation

Quantifying Uncertainty in Deep Spatiotemporal Forecasting

Knowledge-aware Coupled Graph Neural Network for Social Recommendation

Traffic Flow Forecasting with Spatial-Temporal Graph Diffusion Network

Fine-Grained Urban Flow Prediction

Self-Supervised Deep Metric Learning for Pointsets

Incidence and Costs of Personal and Property Crimes in the USA, 2017

STAN: Spatio-Temporal Attention Network for Next Location Recommendation

Learning Dynamics and Heterogeneity of Spatial-Temporal Graph Data for Traffic Forecasting

Pre-Trained Image Processing Transformer

Self-Supervised Text Planning for Paragraph Completion Task

Contrastive Learning with Hard Negative Samples

Adaptive Graph Convolutional Recurrent Network for Traffic Forecasting

One-Shot Neural Architecture Search via Novelty Driven Sampling

Self-Supervised Learning: Generative or Contrastive

Connecting the Dots: Multivariate Time Series Forecasting with Graph Neural Networks

Hierarchically Structured Transformer Networks for Fine-Grained Spatial Event Forecasting

Traffic Flow Prediction via Spatial Temporal Graph Neural Network

Urban big data fusion based on deep learning: An overview

GMAN: A Graph Multi-Attention Network for Traffic Prediction

Deep Dynamic Fusion Network for Traffic Accident Forecasting

Dynamic Hypergraph Neural Networks

Urban Traffic Prediction from Spatio-Temporal Data Using Deep Meta Learning

DeepSTN+: Context-Aware Spatial-Temporal Neural Network for Crowd Flow Prediction in Metropolis

Non-Stationary Model for Crime Rate Inference Using Modern Urban Data

Graph WaveNet for Deep Spatial-Temporal Graph Modeling

Estimating Node Importance in Knowledge Graphs Using Graph Neural Networks

MiST: A Multiview and Multimodal Spatial-Temporal Learning Framework for Citywide Abnormal Event Forecasting

UrbanFM: Inferring Fine-Grained Urban Flows

Revisiting Self-Supervised Visual Representation Learning

DeepCrime: Attentive Hierarchical Recurrent Networks for Crime Prediction

Deep Graph Infomax

Hypergraph Neural Networks

Deep Distributed Fusion Network for Air Quality Prediction

GeoMAN: Multi-level Attention Networks for Geo-sensory Time Series Prediction

FakeDetector: Effective Fake News Detection with Deep Diffusive Neural Network

Revisiting Spatial-Temporal Similarity: A Deep Learning Framework for Traffic Prediction

Modeling Temporal-Spatial Correlations for Crime Prediction

Spatio-temporal Graph Convolutional Neural Network: A Deep Learning Framework for Traffic Forecasting

Diffusion Convolutional Recurrent Neural Network: Data-Driven Traffic Forecasting

Constructions of crime data and criminal statistics: A critical reflection

Multi-resolution Spatial Event Forecasting in Social Media

Deep Spatio-Temporal Residual Networks for Citywide Crowd Flows Prediction

Crime Rate Inference with Big Data

Predicting the Next Location: A Recurrent Model with Spatial and Temporal Contexts

Deep Residual Learning for Image Recognition

Utilizing Real-World Transportation Data for Accurate Traffic Prediction

LIBSVM: A library for support vector machines

Et al

Crime data mining: a general framework and some examples

A Comprehensive Survey on Graph Neural Networks

Deep Learning: A Generic Approach for Extreme Condition Traffic Forecasting

Field of Study

Journal Information

Name

Page

Venue Information

Name

Type

URL

Alternate Names