Geography Reference
In-Depth Information
aspects should become a focus for more active public participation through
additional cooperation. Like other geovisual analytics tools, eXplorer embodies a
form of spatial analysis, the product of which is a level of information processed or
extracted from the ''raw'' point data of WikiCrimes and is in a more digestible and
therefore useful form for law enforcement officers and the public alike. Specifi-
cally, eXplorer was chosen for its facility for quick and easy publication to the
web, relatively uncluttered interface, smooth, fast interaction and ability to display
''stories''
alongside
specific
combinations
of
data.
This
combination
made
eXplorer a suitable match with the anticipated public user population.
Finally, the application of geovisual analytics to the WikiCrimes dataset is
representative of what could be done to online volunteered data in general. There
have been very few such attempts (e.g. Jankowski et al.
2010
; Kisilevich et al.
2012
); most geovisual analytics endeavours focus on official data (e.g. regional
and national statistical data-Jern
2009
,
2010
), top-down collected data as opposed
to the bottom-up nature of volunteered data collection (Sui
2008
).
The next section gives some background on visual analytics and geo-collaboration
before an overview of the dataset then outlining specifics in piping data from
WikiCrimes to eXplorer. Finally, an account of the information gleaned from
WikiCrimes through a geovisual analytics approach will be given at the country scale
at a state level and one degree grid cell level before some concluding statements.
2 Background
2.1 Visual Analytics
We live in a world where data comes to us in voluminous and complex forms and
the tools we have to make sense of this data are immature. Visual analytics
emerged from NVAC (National Visualization and Analytics Center) in the US in
order to address this problem, specifically in the context of preventing/responding
to terrorist attack.
NVAC have defined visual analytics as ''the science of analytical reasoning facil-
itated by interactive visual interfaces'' (Thomas and Cook
2005
, p 4). The mantra here
is to ''detect the expected and discover the unexpected'' from synthesized information
elicited from raw data. This is effected in many ways: supporting the human facility for
analytical reasoning, transforming data into optimal representations for analysis, and
using visual representation and interaction to mine information. The results of a
structured assessment of the data using these groups of technologies then has to be
packaged and disseminated into an easily digestible and actionable form for public and
decision makers alike (Thomas and Cook
2005
).
Geovisual analytics is distinguished mainly by the enhanced complexity of the
geospatial domain (Andrienko et al.
2007
), with facets of scale, multidimensionality
and autocorrelation, to name but a few. Examples of geovisual analytics research
