Global Positioning System Reference
In-Depth Information
features, taking into consideration real-world applications and analyzing
whether there are some common steps for cleaning patterns. This could
help to expand ETL tools by including a (complex) component in charge
of (at least some) cleaning processes, e.g., spatial aggregation of geometries
for objects with the same identifi er, and validation of geometries, among
others. Furthermore, when using the same shape fi le for different DBMSs,
we found that they differ in criteria to check the validity of a geometry.
There are already solutions for transforming invalid geometry into a valid
one that depends on chosen DBMSs and they could also be included as a
component of the spatial ETL tool. By doing so, users are not forced to have
knowledge about buffering, inverting the specifi cation order for coordinates,
or self-intersection, among other aspects.
Spatial OLAP server layer
Multidimensional modeling concepts applied to spatial data have been used
in various spatial OLAP proposals and solutions. Different publications have
considered the integration between GISs and DW or OLAP environments.
Pourabbas (2003) and Ferri et al. (2000) referred to common key elements of
spatial and multidimensional databases: time and space. Based on a formal
model they achieved the integration by applying mapping between the
hierarchical structures of the OLAP and GIS environments. The concept of
this kind of mapping was also exploited by other authors (Escribano et al.
2007; Kouba et al. 2000). Different implementations with server and client
layers, e.g., GeoWOlap (Bimonte et al. 2006), Piet (Escribano et al. 2007),
Map4Decision (Intelli3 2013), SAS Web OLAP Viewer (SAS 2013), SOVAT
(Scotch and Parmanto 2005), GOAL (Kouba et al. 2000), GeoMDCube
(Silva, Manhäes and Gitahy 2008), include a user interface that hides the
fact that two separate systems compose the SOLAP architecture: OLAP
system (e.g., Microsoft SQL Server Analysis Services, Mondrian, GOLAPE,
or SAS Enterprise BI Server) with GIS solutions (e.g., ESRI ArcGIS or
some customized applications that facilitate map display based on Java
MapXtreme or others). The above mentioned SOLAP systems are not always
available to a wide spectrum of users since if they are commercial, e.g., SAS
Web OLAP Viewer (SAS 2013) or Map4Decision (Intelli3 2013), they need
some investment that may be diffi cult to justify for inexperienced users
with little or no knowledge about OLAP. On the other hand, some of the
solutions are prototypes that (1) are not publicly available (to the best of
our knowledge), e.g., GeoCube (Bimonte et al. 2005), GeoWOlap (Bimonte
et al. 2006), Piet (Escribano et al. 2007; Gómez et al. 2011), GOLAPE (Silva
et al. 2006), GeoOlap (Soares et al. 2007), (2) are customized (no general-
purpose solutions), e.g., SOVAT developed for public health decision
Search WWH ::
Custom Search