Chen, T;
Cheng, T;
(2020)
Identifying Optimal Scales for Spatio-temporal Crime Clusters.
In:
Proceedings of the 28th Geographical Information Science Research conference: GISRUK 2020.
(pp. p. 91).
GISRUK
Preview |
Text
GISRUKPaper2020.pdf - Published Version Download (852kB) | Preview |
Abstract
The spatial and temporal scales are not only two essential parameters for the spatio-temporal clustering algorithm to generate the crime clusters but are significantly helpful for determining the interventive distance at space and time in place-based crime prevention. This study presents the issue of identifying the optimal spatial-temporal scale when examining the micro-level crime clusters approached by density-based spatio-temporal clustering methods. The approach comprises adopting a clustering evaluation index to examine the performance of different clustering results from a range of space and time values iteration. For this purpose, two types of density-based clustering algorithms called ST-DBSCAN and ST-OPTICS are compared to determine the optimal scales for space-time crime clusters. A case study is demonstrated using individual crime records of burglary from Vancouver, Canada in 2010. Several derived results are significant. First, appropriate scales – 500m and 3 days can be distinctively determined by clustering algorithm ST-OPTICS from our tested parameters. Second, the narrowed scales were found in this study significantly for spatio-temporal crime clusters, which can help to develop a more focused and specific policing tactics.
Type: | Proceedings paper |
---|---|
Title: | Identifying Optimal Scales for Spatio-temporal Crime Clusters |
Event: | Geographical Information Science Research- UK (GISRUK) 2020 |
Location: | London |
Dates: | 21 July 2020 - 23 July 2020 |
Open access status: | An open access version is available from UCL Discovery |
Publisher version: | http://london.gisruk.org/gisruk2020_proceedings/GI... |
Language: | English |
Additional information: | This version is the version of record. For information on re-use, please refer to the publisher’s terms and conditions. |
Keywords: | Spatio-temporal clustering, crime hotspot, near-repeat victimisation, density-based, clustering, ST-OPTICS |
UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Civil, Environ and Geomatic Eng |
URI: | https://discovery.ucl.ac.uk/id/eprint/10114785 |
Archive Staff Only
View Item |