DBSCAN clustering method is applied to identify severe Traffic Accident (TA) hotpots on roads

  • Arkadiy Gershtein State University, 7–9, Universitetskaya emb., 198504, Saint Petersburg, Starii Petergof, Russia.
  • Andrey Terekhov Saint Petersburg State University, 28 Universitetskiy pr., Stary Peterhof, 198504, Saint Petersburg, Russia
Keywords: vehicle traffic accident hotspot cluster DBSCAN simulation Monte-Carlo Massachusetts

Abstract

DBSCAN clustering method is applied to identify severe Traffic Accident (TA) hotpots on roads. The research examines severe TA, defined as those that led to human damage (injury or death), in the city of Newton, MA and in the entire state of Massachusetts, USA from 2013 to 2018. DBSCAN algorithm was also applied to network-constrained uniformly distributed over road network data to locate threshold in number of points per cluster so that all more populated clusters identified in real data can be treated as statistically significant. For DBSCAN algorithm two types of distance metrics, Euclidean and over Network, were compared. It is found that both distances are equivalent on scale of 10 meters, which justifies hybrid approach to clustering: using Network distance only to generate uniformly distributed points needed for Monte-Carlo simulations. All clustering can be performed using Euclidean distances which is much faster and more memory efficient. Subsequent years analysis demonstrates the extend that hotspots identified are stable and occur consecutively for several years and hence may possess predictive value.

Author Biographies

Arkadiy Gershtein, State University, 7–9, Universitetskaya emb., 198504, Saint Petersburg, Starii Petergof, Russia.

Postgraduate of the Faculty of Mathematics and Mechanics, SPbSU, ArkadyGer@gmail.com

Andrey Terekhov, Saint Petersburg State University, 28 Universitetskiy pr., Stary Peterhof, 198504, Saint Petersburg, Russia

PhD, Professor, Head of the Department of System Programming of the Faculty of Mathematics and Mechanics, SPbSU, a.terekhov@spbu.ru

References

S. Chainey, L. Tompson, and S. Uhlig, “The utility of hotspot mapping for predicting spatial patterns of crime,” Secur J, vol. 21, no. 1, pp. 4–28, 2008; doi: 10.1057/palgrave.sj.8350066

S. Chainey and J. Ratcliffe, GIS and Crime Mapping, Chichester, UK: John Wiley and Sons, 2005; doi: 10.1002/9781118685181

M. Ester, H-P. Kriegel, J. Sander, and X. Xu, “A density-based algorithm for discovering clusters in large spatial databases with noise,” in Proc. KDD’96: 2nd Int. Conf. on Knowledge Discovery and Data Mining, 1996, pp. 226–231.

A. Gramacki, Nonparametric Kernel Density Estimation and Its Computational Aspects, Cham, Swi- tzerland: Springer International Publishing, 2018; doi: 10.1007/978-3-319-71688-6

P. A. P. Moran, “Notes on Continuous Stochastic Phenomena,” Biometrika, vol. 37, no. 1/2, pp. 17–23, 1950; doi: 10.2307/2332142

A. Okabe and K. Sugihara, Spatial Analysis along Networks: Statistical and Computational Methods, Hoboken, NJ, USA: John Wiley & Sons, 2012; doi:10.1002/9781119967101

P. Songchitruksa and X. Zeng, “Getis–Ord Spatial Statistics to Identify Hot Spots by Using Incident Management Data,” Transportation Research Record, no. 2165, pp. 42–51, 2010; doi: 10.3141/2165-05

L. Yingjie, et al., “Mapping the hotspots and coldspots of ecosystem services in conservation priority setting,” Journal of Geographical Sciences, vol. 27, no. 6, pp. 681–696, 2017; doi: 10.1007/s11442-017- 1400-x

Y. Xie and S. Shekhar, “Significant DBSCAN towards Statistically Robust Clustering,” in Proc. SSTD’19: 16th International Symposium on Spatial and Temporal Databases, 2019, pp. 31–40; doi: 10.1145/3340964.3340968

“Massgis data-massachusetts department transportation massdot roads,” in docs.digital.mass.gov, [Online]. Available: https://docs.digital.mass.gov/dataset/massgis-data-massachusetts-department-transportation-massdot-roads

“QGIS,” in www.qgis.org, [Online]. Available: https://www.qgis.org/en/site/

“Open jump,” in www.openjump.org, [Online]. Available: http://www.openjump.org/

“MassDOT Crash Open Data Portal,” in Mass.gov, [Online]. Available: https:// massdot-impact-crashes-vhb.opendata.arcgis.com/search

“SANET,” in sanet.csis.u-tokyo.ac.jp, [Online]. Available: http://sanet.csis.u-tokyo.ac.jp/

Published
2021-03-28
How to Cite
Gershtein, A., & Terekhov, A. (2021). DBSCAN clustering method is applied to identify severe Traffic Accident (TA) hotpots on roads. Computer Tools in Education, (1), 45-57. https://doi.org/10.32603/2071-2340-2020-46-58
Section
Informational systems