VisCrimePredict: a system for crime trajectory prediction and visualisation from heterogeneous data sources

Open multidimensional data from social media and similar sources often carries insightful information on social issues. With the increase of high volume data and the proliferation of visual analytics platforms, it becomes easier for users to interact with and select meaningful information from large data sets. The prevention of crime is a crucial issue for law-enforcing agencies tasked with maintaining societal stability. The ability to visualise crime patterns and predict imminent incidents accurately opens new possibilities in crime prevention. In this paper, we present VisCrimePredict, a system that uses visual and predictive analytics to map out crimes that occurred in a region/neighbourhood. VisCrimePredict is underpinned by a novel algorithm that creates trajectories from heterogeneous data sources such as open data and social media with the aim to report incidents of crime. VisCrimePredict uses a Long Short Term Memory (LSTM) algorithm for trajectory prediction. A proof of concept implementation of VisCrimePredict and an experimental evaluation of crime trajectory prediction accuracy using LSTM neural network concludes the paper.

[1]  A. Agresti,et al.  Statistical Methods for the Social Sciences , 1979 .

[2]  Prem Prakash Jayaraman,et al.  VisCrime: A Crime Visualisation System for Crime Trajectory from Multi-Dimensional Sources , 2019, WSDM.

[3]  Andrea Cavallaro,et al.  Global trajectory reconstruction from distributed visual sensors , 2008, 2008 Second ACM/IEEE International Conference on Distributed Smart Cameras.

[4]  W. Tobler.  A Computer Movie Simulating Urban Growth in the Detroit Region , 1970 .

[5]  Yu Zheng,et al.  Trajectory Data Mining , 2015, ACM Trans. Intell. Syst. Technol..

[6]  Jürgen Schmidhuber,et al.  Connectionist temporal classification: labelling unsegmented sequence data with recurrent neural networks , 2006, ICML.

[7]  Peng Chen,et al.  Forecasting Crime Using the ARIMA Model , 2008, 2008 Fifth International Conference on Fuzzy Systems and Knowledge Discovery.

[8]  Stan Matwin,et al.  A Semi-Supervised Approach for the Semantic Segmentation of Trajectories , 2018, 2018 19th IEEE International Conference on Mobile Data Management (MDM).

[9]  S. Loke,et al.  Here-n-Now : A Framework for Context-Aware Mobile , 2012 .

[10]  Reda Alhajj,et al.  Crime Hot-Spots Prediction Using Support Vector Machine , 2006, IEEE International Conference on Computer Systems and Applications, 2006..

[11]  Zhifeng Bao,et al.  Supporting Large-scale Geographical Visualization in a Multi-granularity Way , 2018, WSDM.

[12]  Matthew S. Gerber,et al.  Predicting crime using Twitter and kernel density estimation , 2014, Decis. Support Syst..

[13]  Klaus Schulten,et al.  Fast Visualization of Gaussian Density Surfaces for Molecular Dynamics and Particle System Trajectories , 2012, EuroVis.

[14]  Okran Jeong,et al.  A trajectory analysis system for social media contents using AsterixDB , 2017, IMCOM.

[15]  Prem Prakash Jayaraman,et al.  Visual Analytics Ontology-Guided I-DE System: A Case Study of Head and Neck Cancer in Australia , 2018, 2018 IEEE 4th International Conference on Collaboration and Internet Computing (CIC).

[16]  Michael Leitner,et al.  Population at risk: using areal interpolation and Twitter messages to create population models for burglaries and robberies , 2017, Cartography and geographic information science.

[17]  Wei Ding,et al.  Crime Hot Spot Forecasting: A Recurrent Model with Spatial and Temporal Information , 2017, 2017 IEEE International Conference on Big Knowledge (ICBK).

[18]  Xiaoli Wang,et al.  Noise filtering, trajectory compression and trajectory segmentation on GPS data , 2016, 2016 11th International Conference on Computer Science & Education (ICCSE).

[19]  Martin A. Andresen,et al.  The Trajectories of Crime at Places: Understanding the Patterns of Disaggregated Crime Types , 2017 .

[20]  S. Bushway,et al.  Trajectories of Crime at Places: A Longitudinal Study of Street Segments in the City of Seattle , 2004 .

[21]  David C. Hogg,et al.  Learning the Distribution of Object Trajectories for Event Recognition , 1995, BMVC.

[22]  Prem Prakash Jayaraman,et al.  Context-Aware Recruitment Scheme for Opportunistic Mobile Crowdsensing , 2015, 2015 IEEE 21st International Conference on Parallel and Distributed Systems (ICPADS).

[23]  Mihai Surdeanu,et al.  The Stanford CoreNLP Natural Language Processing Toolkit , 2014, ACL.

[24]  Jürgen Schmidhuber,et al.  Long Short-Term Memory , 1997, Neural Computation.

[25]  Martin A. Andresen.  Environmental Criminology: Evolution, Theory, and Practice , 2014 .

[26]  Maike Buchin,et al.  Segmenting trajectories: A framework and algorithms using spatiotemporal criteria , 2011, J. Spatial Inf. Sci..

[27]  Xinyu Chen,et al.  Crime prediction using Twitter sentiment and weather , 2015, 2015 Systems and Information Engineering Design Symposium.