Geoscience Reference
In-Depth Information
Observations contributed by volunteers or end-users take many forms, but
generally end up translated into text, which is geoparsed for relevant
placenames, directions, distances, and geographically-relevant prepositions
and prepositional phrases, such as 'nearby', 'next to', and 'on top of'.
Observations can also be obtained using a technology such as GeoRSS
(Geographically Encoded Objects for Really Simple Syndication).
GeoRSS is a process of extracting geographical information, such as
latitude and longitude, from any geographically tagged feed. These feeds
are useful when one wants to keep track of regularly changing information
such as news and traffic conditions. There are two encodings of GeoRSS.
GeoRSS-Simple is the simplified version of encoding whereas GeoRSS
Geographic Markup Language (GML) has more features, including the
availability of more than one specific coordinate system. In order to extract
geographical data from social network feeds such as Twitter and Face-
book, the locational information has to be in the feed itself. This requires
third party software since both aforementioned applications are still develop-
ing geo-tagging on their formats. Much controversy exists over privacy
concerns as well as location accuracy, but GeoRSS and related geotech-
nologies form an important aspect of the observation collection process,
and we have developed some procedures for masking sensitive or private
information associated with end-users that contribute information.
After observations are collected, a list of relevant placenames is obtained
from our localized gazetteer, which contains a comprehensive list of official
campus feature names, colloquial variants of placenames, abbreviations,
coded placenames, and common foreign-language variants of placenames.
The observations are then associated with spatial footprints of the features
from the localized gazetteer and plotted on a map.
The Geoparsing software tool being developed for this project uses
VB.NET together with MapWindow GIS OCX (Object Linking and
Embedding Control eXtension) control. MapWindow GIS is an open source
GIS application under the Mozilla Public License, started at Utah State
University, Logan, Utah (Ames et al. 2008). Since MapWindow GIS is an
open source software package, GIS programmers are permitted to config-
ure, use, and improve the software code for their specific needs. There are
two GIS programming paradigms in MapWindow GIS: standalone appli-
cations and plug-ins (Aburizaiza and Ames 2009). For this project, the
standalone development approach was utilized. A screenshot of the
