Global Positioning System Reference
In-Depth Information
fact, spatial similarity has been defi ned as matching and ranking according
to a certain context (function, goal), scale (coarse or fi ner level), and
technology (for searching, retrieving, and recognizing data). The research
on spatial similarity ranges from data retrieval (Papadias and Delis 1997;
Samet 2004), problem solving (Yeh and Shi 2003), confl ation (Cobb et al. 1998),
to interoperability (Harvey 1999), etc.
The importance of semantic similarity of geographical concepts in
GIScience has been highly emphasized by tasks of sharing and interoperation
of geographical information (Buccella et al. 2011). These tasks are addressed
to support the sharing and reuse of formally represented knowledge
among GISs and are supported through a common ontology in which
shared knowledge is represented. Typically, ontologies can be regarded
as general tools of information representation on a specifi c subject. In fact,
there are several examples of using ontologies in the spatial data domain.
In the Spatial Data Transfer Standard (SDTS), 1 for instance, an ontology has
been used to describe the underlying conceptual model and the detailed
specifi cations for the content, structure, and format of spatial data, their
features and associated attributes. In SDTS, concepts are commonly used
on topographic quadrangle maps and hydrographic charts (USGS 2012).
The GeoNames 2 ontology makes it possible to add geospatial semantic
information to the Web as well. GeoNames is a community-driven database
which contains place names and contains 8 million entries. It provides
services to describe the relation between toponyms or place names. A
further example is LinkedGeoData 3 that has been emerged to enhance
OpenStreetMap 4 semantics. LinkedGeoData (LGD) has taken the entire
OpenStreetMap data set and linked it to a formal ontology, which is a
simple and shallow tree structure, representing keys and values (Ballatore
et al. 2012). Its semantic content is limited to Is-A relationships between
tags and respective values.
Principally, semantic similarity measures are necessary to quantify
similarities of concepts belonging to the same or different ontologies (Gruber
1993; Visser et al. 2002), and are applied to a wide range of applications
including the integration of geographic information from different sources,
data mining, semantically enabled gazetteer services, semantics based
geographic information search and retrieval as well as in conjunction with
any Web search engine for enhancing the performance of retrieval (Jones
et al. 2003).
1
http://mcmcweb.er.usgs.gov/sdts/
http://www.geonames.org/
2
3
http://linkedgeodata.org/
http://www.openstreetmap.org/
4
Search WWH ::




Custom Search