Spatio-temporal mixed membership models for criminal activity
Published version
Peer-reviewed
Repository URI
Repository DOI
Change log
Authors
Virtanen, S
Girolami, Mark https://orcid.org/0000-0003-3008-253X
Abstract
jats:titleAbstract</jats:title> jats:pWe suggest a probabilistic approach to study crime data in London and highlight the benefits of defining a statistical joint crime distribution model which provides insights into urban criminal activity. This is achieved by developing a hierarchical mixture model for observations, crime occurrences over a geographical study area, that are grouped according to multiple time stamps and crime categories. The mixture components correspond to spatial crime distributions over the study area and the goal is to infer, based on the observations, how and to what degree the latent distributions are shared across the groups.</jats:p>
Description
Keywords
Bayesian statistics, high-dimensional data, latent factor models, multi-view modelling, spatial and temporal methods
Journal Title
Journal of the Royal Statistical Society. Series A: Statistics in Society
Conference Name
Journal ISSN
0964-1998
1467-985X
1467-985X
Volume Title
184
Publisher
Oxford University Press (OUP)
Publisher DOI
Sponsorship
EPSRC (EP/P020720/2)
EPSRC (EP/R018413/2)
Engineering and Physical Sciences Research Council (EP/R034710/1)
Royal Academy of Engineering (RAEng) (RCSRF\1718\6\34)
EPSRC (EP/R018413/2)
Engineering and Physical Sciences Research Council (EP/R034710/1)
Royal Academy of Engineering (RAEng) (RCSRF\1718\6\34)