The First International Workshop on Managing Data with Mobile Devices

A MOBILE GEOGRAPHIC INFORMATION SYSTEM MANAGING

CONTEXT-AWARE INFORMATION BASED ON ONTOLOGIES

Anderson Resende Lamas, Jugurta Lisboa Filho,

Alcione de Paiva Oliveira, Ronoel Matos de Almeida Botelho Jnior
Federal University of Viosa Departamento de Informtica
Viosa, MG, Brazil CEP: 36570-000
{andersonlamas,jugurta}@ufv.br,{alcione, ronoeljr}@dpi.ufv.br

ABSTRACT
This paper presents the development of a Mobile Geographical Information System
(Mobile GIS) capable of managing context information. This system was established
from an architecture based on the specification of an ontology-based context model and a
set of Web Services to access information remotely stored in a geographic database. This
mechanism allows Mobile GIS users to receive personalized information in their mobile
devices, combining the information on their profiles with the display of geo-spatial data.

Keywords: Mobile geographic information systems, context-aware information.

1 INTRODUCTION device from storing a large amount of data locally or

A number of fields of knowledge have been
expanded through the use of technologies that exploit
mobility and communication between users and
devices. The steady expansion of the infrastructure
wireless network and high proliferation of mobile de-
vices, such as PDA `s (Personal Digital Assistant),
Smartphones and mobile phones are some of the
technologies that have favored this scenario.
In the field of Context-Aware Applications, for
instance, one can use the dynamic context of mobile
devices users, caused by the mobility and constant
change in environment, and then to delivery
personalized information. Context sensitive
applications are characterized by use of context
information to provide services and information
relevant to users during a task execution [3]. In this
scenario, context can be defined as any information
that can be used to characterize a person, a place or
object, relevant to an interaction between a user and
an application [3].
As for the field of Geographic Information
Systems (GIS), the intersection of geo-spatial data
with mobile devices has led to the field of Mobile
GIS. According to [8], Mobile GIS can be defined as
a framework that integrates hardware and software to
access spatial data (maps) through a wireless
network, using mobile devices.
Considering their peculiar characteristics, both
Mobile GIS applications and context-aware mobile
applications bring with them major challenges to
their development. A common challenge to these
areas is to create a mechanism to bypass the limited
capacity of mobile devices. Among these limitations
are the little space to store data, low processing
power, small screens and low bandwidth of wireless
network [18]. These factors prevent, for example, the
even processing them and give a timely reply.
For Mobile GIS, there is still the challenge of
using proper technologies for storage, query and
display of geo-spatial data on mobile devices. As
example of these technologies we can mention
geographic databases [19] and Geo Web Services
[20].
For context-aware applications, one the of the
challenges is how to define the most comprehensive
and appropriate context model for the application
domain being developed. Thanks to the power of
representation and formal specification of ontologies,
many studies have proposed this approach as a way
of modeling context [14], [4] and [12].
The integration of context information in a
Mobile GIS environment can add great value to the
information that is displayed to an application user.
This is due to the possibility of delivering
personalized information to the user in combination
with the display of geo-spatial data. In this scenario,
this paper presents the characteristics and existing
solutions for both the field of Mobile GIS and
context sensitive applications and proposes an
architecture for developing applications of Mobile
GIS capable of managing context information.
Section 2 presents the architecture proposed for
the development of context sensitive Mobile GIS.
Section 3 presents a case study built to validate this
architecture. Section 4 describes some related works.
Finally, Section 5 presents the conclusions of this
work.
2 THE CM-GIS ARCHITECTURE
(CONTEXT-MOBILE-GIS)
The architecture proposed in this paper, named
Context-Mobile-GIS (CM-GIS) aims to construct

UbiCC Journal Volume 4 718

The First International Workshop on Managing Data with Mobile Devices

integrated context sensitive mobile GIS applications.

The development of Mobile GIS applications, as
defined by Tomko [10], requires attention to: (a)
storage and data query in a spatial database; (b)
provide a mechanism for querying and accessing
data remotely located; (c) the correct use of
technology.
Regarding the geographic database (a) one must
use a Database Management Sys-tem (DBMS) that is
capable of storing and managing geo-spatial data.
Some DBMS available today provide this
functionality through an extension for spatial data,
such as Oracle, SQLServer, PostGreSQL, among
others. To deal with remote access to information (b), Figure 1: The SeCoM Model [1].
a frequently adopted solution is the use of Web
Based on the above-made considerations, we
Services, because of their ability to provide services
that can be dynamically accessed by a network. proposed the CM-GIS architecture for developing
Finally, with respect to the adopted technologies (c), applications of context-sensitive Mobile GIS, as in
Fig. 2. This architecture requires the following
it is important to choose a tool capable to manipulate
elements for the development of mobile GIS
and display geo-spatial data, both in its graphic and
descriptive form in mobile devices. applications able to manage context information:
In the case of context sensitive applications is 1. the mobile GIS application, which will be the
interface for the clients from their mobile
important to establish a mechanism capable of
devices. An important requirement for this
representing and manipulating the context involved
in the domain being treated by the application. application is to be able to manipulate and
Strang and Linnhoff-Popien [9] discuss that the key display geo-spatial data;
2. a set of Web Services to manage the
to the development of a context sensitive application
communication and provision of services and
is the definition of an appropriate context model,
which can be defined by using ontologies. information to the user. The use of Web
The ontological characteristics of formality, Services has a great importance in this ar-
chitecture, by allowing that operations
explicit semantic and implementation abstraction
demanding a high load of processing can be
enable software systems not only to infer new
information from ontology-modeled information, but performed on more robust machines and not
also to share the information among themselves so as directly on the mobile device;
3. a module that communicates with Web Services
to integrate, in a transparent way, the services that
and performs spatial queries, via SQL language,
handle it [1].
Based on these characteristics, Bulco Neto [1] on information stored in geographic database;
proposed the Semantic Context Model (SeCoM), a 4. a set of ontologies used to model the context
involved in the application. We suggest using
domain-independent ontological model described in
the SeCoM model for modeling the context;
Web Ontology Language (OWL), which serves as
basis for the definition of information in context. 5. a defined module from the context model based
This model consists of a set of inter-related on ontologies that communicates with the Web
ontologies based on semantic dimensions of identity, Services to provide information in context. The
information in con-text should also be stored in
location, time and activity.
SeCoM was developed using generic concepts in the database and queried via SQL;
a way that it can be reused for a variety of context 6. a geographic database that manages all the geo-
spatial data and descriptors in-volved in the
sensitive applications. To make this possible, the
application.
ontologies follow a two-layer approach: a top layer
of ontologies, shown in Fig. 1, representing the These items must be detailed according to the
model itself and the bottom layer of ontologies, application that will be developed. The next section
presents a case study, built upon this architecture.
which is built with concepts specific to the context-
sensitive application that is being developed. In this
layer, the SeCoM model can be reused or even
extended with the knowledge specific to the
application domain. Since the SeCoM model was
generically defined, in this paper we suggest its use
in the definition of the context model.

UbiCC Journal Volume 4 719

The First International Workshop on Managing Data with Mobile Devices
Figure 2: Context-Mobile-GIS Architecture (CM-
GIS).
3 UFV-GEOMOBILE: A CASE STUDY
This section describes the development of the
UFV-GeoMobile system. This system was to target
the campus of the Federal University of Viosa
(UFV), which is a public institution of higher
education, with the main campus located in the city
of Viosa, Minas Gerais State. The university offers
several under-graduate and graduate courses in
numerous fields of knowledge. It receives annually a
large number of people visiting or establishing some
kind of relationship with its campus. Among these
people, there are new students (~ 2000 per year),
conference participants, visiting researchers, or
simply people who want to visit their facilities.
Moreover, there are also people who already have
some connection with the University, such as
professors, clerks or students. The common feature
among these people is their often difficulty in
identify-ing sites or services in campus, given its
large area and the appearance of new buildings and
facilities.
The universitys present scenario seems to be
ideal for implementing a system to validate the CM-
GIS architecture proposed in this work. The
existence of a practical and automated information
source available to the user seeking guidance in the
cam-pus is a necessity. It would be even more
appropriate that the options for querying and
obtaining the information were customized according
to the context in which the user is. For example, a
visitor is probably not interested in the time of the
next class, but on the directions of campus services
(bank, pharmacy, etc.) or even the place of any event.
The UFV-GeoMobile was implemented,
allowing a user of a Mobile GIS application on PDA
to query or receive personalized information about
the UFV Campus. This application is based on a map
UbiCC Journal Volume 4
of the campus properly managed by a component for
geographic data handling. In addition, it has a set of
query screens and a mecha-nism for information
display based on the user context. More precisely,
the context definition is based on a calendar of
activities to be performed by the user and so to
display the geographic location of the place where
this activity will be performed.
The whole process of development is described
in the following sections.
3.1 The context model: ontology of a university
campus
The context model for the case study UFV-
GeoMobile must contain information on the structure
and activities carried out in a university campus.
Information that must be modeled to reflect this
domain includes:
People: in this case students, staff and visitors.
Employees may be teachers, researchers or
technical and administrative staff;
The administrative structure: it reflects the
hierarchical organization of a university, including
rectory, centers of education, departments and
councils;
The academic structure: involves activities related
to teaching, research and extension;
Calendar of activities: is the set of tasks that a
person can perform on campus. This item is
essential for the system to consider context
information;
Geo-spatial data: used to represent spatially the
elements with geographic location on campus.
Examples of these elements are buildings and
services installed in each of these buildings.
From this information it is initiated a formal
representation of a context model. The context model
of the UFV-GeoMobile system was based on the
SeCoM model, proposed by Bulco Neto [1], using
the ontological concepts Actor, Activity, Spatial
Event and Temporal Event.
A domain ontology defined as context model
was termed OntoUFV and is represented in Figure 3.
In this ontology, it was defined that people, derived
from the ontology Actor of the SeCoM model, are
divided into Visitor, Student and Staff, with the last
item divided into Technician and Professor. These
people are involved in activities inherited from the
ontology Activity that occurs within a time interval
(TimeInterval) and in a particular location (Physical
Location).
The locations may be buildings or facilities,
which are contained within the buildings. A location
may be the campus itself, which has an
Administrative Structure and an Academic Structure.
The Administrative Structure consists of a set of
Councils, which are divided into Technical
Administrative Councils, Central Administration
Councils and Representative and Adjoining Councils.
720
The First International Workshop on Managing Data with Mobile Devices
The Central Administration Councils are divided into
Rector, Pro-Rectors, Centers of Education and
Departments. The Rector governs the Pro-Rectors
and the Centers of Education. The Pro-Rectors
govern the Technical Administrative Bodies and the
Centers of Education govern the Departments.
The Academic Structure is divided into a number
of Fields of Knowledge, which have different
courses. Each course has a set of subjects which are
offered by departments.
The context model has great importance to
develop the application, as it was used as basis for
the modeling of geographic database and the
Figure 3: OntoUFV: context model of a university campus.
to Shapefile format, using the Cad2shape software
[2]. The following layers were produced
thoroughfare, buildings and specific facilities within
a building, such as administrative offices and
UbiCC Journal Volume 4
implementation of features available in the context
management module.
3.2 Building the geographic database
The geographic database of the UFV-GeoMobile
system was built from the aggregation of a geospatial
data set, kept by the university administration,
associated with a set of relational databases used by a
variety of administrative systems, also maintained by
the institution. The geographic data have been
modified to fit formats of data handled by the Mobile
GIS application. Initially, data were stored in the
format of the AutoCAD system and were converted:
laboratories.
The conceptual modeling of the database, as
shown in Fig. 4, was determined from the
information modeled in the domain ontology of a
721
The First International Workshop on Managing Data with Mobile Devices
university campus. For this task, we used the UML-
GeoFrame model [6], which is a specific model for
geo-graphic database. Subsequently, the conceptual
model was implemented using DBMS PostGreSQL
with its spatial PostGIS extension.
The class diagram of Fig. 4 shows that the
system has classes with and without spatial
representation. According to the UML-GeoFrame
model, classes and sub-classes Clerck, Professor,
Admin Technician, Person, Student, Visitor,
Calendar, Activity, Council and Type of Facility are
specializations of non-geographic objects ( ), i.e.,
with no spatial representation. The classes Building,
Thoroughfare and Facilities are perceived in object
view ( ) and have spatial representation of the type
Figure 4: Modeling of Geographic Database for UFV-GeoMobile.
display geo-spatial data, which is a set of controls
based on the platform .NET to build GIS applications
in mobile devices [7].
To allow access to information stored in the
geographic database a Web Service was
implemented in Java (SOAP), using NetBeans IDE
6.1 and the Glassfish server. For every feature
available in the application of Mobile GIS, a
corresponding operation in the Web Service was
UbiCC Journal Volume 4
Area, Line and Point, respectively. Finally, the
classes AerialView and ContourLine are perceived in
the field view ( ). The first has spatial
representation of the type GridCells ( ), whereas
the second has representation of the type Isolines
( ).
3.3 Mobile GIS Application
The graphical interface of Mobile GIS was
developed using the framework Microsoft Visual
Studio 2005, C#.NET language and the Pocket PC
2003 emulator. SharpMap software was used to
developed to receive the parameters and pass them to
the Spatial Queries Module or to the Context
Management Module. To illustrate these operations,
this work describes the option of locating a facility or
service, location a person and the mechanism that
shows the geographic location of an activity to be
performed.
To locate a service or facility, initially the user
enters the full or part of the name of the desired item,
722
The First International Workshop on Managing Data with Mobile Devices
using the screen shown in Fig. 5-a. When typing the
name and clicking on Search, the GetLocation
operation is called, as described in Table 1.
This operation receives as parameters the typed
name and the users coordinates (x,y) via GPS or via
a mechanism implemented in Mobile GIS, which
allows the users to indicate their location by inserting
an icon on the map. This second choice was
implemented so the user can use the UFV
GeoMobile application, even with a PDA without an
embedded GPS.
The operation GetLocation passes its parameters
to the query module, generating a SQL query, also
described in Table 1, which returns the names of
sites that match the name entered, the distance of
each facility in relation to the user's position and the
coordinates of the buildings center point in which
the facility is allocated. The selection carried out by
the SQL query is returned to the mobile device that
displays a list of available facilities (Fig. 5-a). By
selecting one of the items in this list, the application
indicates, through a small icon on the map, the
location of the chosen facility, as shown in Fig. 5-b.
The operation to locate a person is similar to the
process for identifying a service or facility. The
location of a person is defined by the association of
the individual with her or his work within the UFV.
Initially the user enters the full or part of the persons
(a) Facility query
Figure 5: Query screens for facilities in the UFV-GeoMobile system.
UbiCC Journal Volume 4
name using the appropriate screen. Clicking the
Search button generates the operation GetPerson of
the Web Service, which is detailed in Table 2. This
operation generates the parameters name and
coordinates (x, y) of the user.
The operation GetPerson passes its parameters to
the Spatial Query Module, generating an SQL query
(Table 2), returning peoples names which
correspond to the entered name, the respective
UFVs council to which the person is linked, the
distance of each council in relation to the user's
position and the coordinates of the center point of the
building in which the council is allocated. The
selection made by the SQL query is returned to the
mobile device, which displays a list of possible
people who meet the entered name. By selecting one
of the items in this list, the application indicates, via
a small icon, the location of the council to which the
chosen person is linked.
The mechanism to notify the Mobile GIS users
on the geographic location of any activity registered
in their diaries was developed as a sensitive choice to
the users context. Initially, to have access to this
feature the user must be identified in the system
using a login and password. From this moment,
every minute the operation GetEvent is called, as in
(b) Facillity location.
723
The First International Workshop on Managing Data with Mobile Devices

Table 1: Web Service operation and SQL query to locate facilities

.
@WebMethod(operationName = "GetLocation")
public List<String> GetLocation(
@WebParam(name = "LocationName") String LocationName,
@WebParam(name = "x") double x,
@WebParam(name = "y") double y) {
.....
}

SELECT f.name, distance(b.the_geom,

GeomFromText('POINT(x, y)',-1)), Centroid(b.the_geom)
FROM Building b, Facility f, FacilityType ft
WHERE (f.name like '%LocationName%') or
((ft.type= 'LocationName') and (ft.type = f.type)) and
intersects(b.the_geom, f.the_geom);

Table 2: Web Service operation and SQL query to locate persons.

@WebMethod(operationName = "GetPerson")
public List<String> getPerson(@WebParam(name = "PersonName")
String PersonName, @WebParam(name = "x")
double x, @WebParam(name = "y")
double y) {
...
}

SELECT cl.nome, co.nome,

distance(b.the_geom, GeomFromText('POINT(x, y)',-1)),
Centroid(b.the_geom)
FROM Clerk cl, Council co, Building b, Facility f
WHERE (cl.nome like '%PersonName%') and
intersects(b.the_geom, f.the_geom) and
(co.idCouncil = cl.idCouncil) and
(co.idCouncil = f.idCouncil)

Table 3: Web Service operation and SQL query to notify events.

@WebMethod(operationName = "GetEvent")
public List<String> GetEvent(@WebParam(name = "login")
String login, @WebParam(name = "password")
String password, @WebParam(name = "actualTime")
Object actualTime) {
....
}

SELECT a.description, a.iniTime, act.endTime,

f.name, Centroid(b.the_geom)
FROM Building e, Facility f, Activity a,
Calendar c, Person p
WHERE (a.horaIni 'actualTime') = 10 and
(p.login = 'login') and (p.password = 'password') and
intersects(b.the_geom, f.the_geom) and
(p.idperson = c.idperson) and
(c.idCalendar = a.idCalendar);

UbiCC Journal Volume 4 724

The First International Workshop on Managing Data with Mobile Devices
Figure 6: Screen to notify the user the occurrence
of any activity.
Table 3. This operation receives as parameter,
besides the login and password, the current time,
which is obtained from the mobile device. This
operation passes these parameters to the context
management module which generates a SQL query
(Table 3). This query returns all the users recorded
activities to start in 10 minutes, describing the
activity in detail, initiation and completion time,
name of the facility in which the activity will be
performed and the coordinates of the buildings
center point where the facility is allocated. From
this information, the Mobile GIS application
displays a text box and indicates to the user the
activity location through an icon, as in Fig. 6.
4 RELATED WORKS
The development of context sensitive mobile
GIS applications has been studied by several groups.
This section describes some of the works that
contributed to accomplish the architecture proposed
here.
Li et al. [5] provided a mechanism based on
Web Services to display contextual information to
mobile device users. Contextual information is
defined through the user's location, via GPS, and
information of used hardware and software. This
context definition does not consider information on
the user's profile or even addresses geo-spatial data.
Weibenberg [12] described the FLAME2008
UbiCC Journal Volume 4
platform, which was developed to be used in the
Beijing Olympics. This platform is featured as a
solution to deliver personalized information to
mobile device users. This work uses a set of
ontologies to define the user situation, trying to
infer the activities that users could carry out over a
period of time. Starting from the definition of the
users position, the most appropriate information
for that time is displayed via Web Services. Despite
being a very wide system, this work does not
describe the implications of integrating information
in context into a Mobile GIS environment.
The CoMPASS architecture (Combining
Mobile Personalized Applications with Spatial
Services) [11] uses, in addition to information on
profile and location, the users trajectory to
automatically deliver personalized information to a
GIS-Mobile client. The definition and storage of
context information are based on the use of a set of
log files. This approach differs from the solution
proposed in this article, which suggests the use of
ontologies to define context information and the use
of a geographic database for storing data.
Brisaboa et al. [13] specifies a middleware to
facilitate the queries on a geographic database from
a mobile device. Access to information is made by
Geo Web Services following the standards
recommended by the OGC. This work provides a
good reference on how to access and manage geo-
spatial data from mobile devices, but does not take
into account the context of the user to display the
information.
Another very interesting work is the project
MyCampus [14] in which a structure was
developed based on Web Services and ontologies
for the students of Carnegie Melon University
(USA) to get on their PDAs a set of information
based on their context within the campus. This
work, although using some information on geo-
spatial data, does not specify the challenges nor the
necessary tools to build applications that handle
such data.
NAMGIS [15] is a open-source context-aware
Mobile Web GIS that exploits the knowledge of the
user position to adapt the interface and track the
user while he moves on the territory shown by the
system on mobile device. It is composed of Core
and a Framework (SAF). The Core provides an
context-adaptable Web interface supporting
principal GIS functionalities, such as map browsing,
feature and attribute based queries. SAF supports
the acquisition and elaboration of context data,
including GPS data, used to track the user's position
in real time and adapt the contrast/brilliance level of
the interface in outdoor scenarios, and RFID (Radio
Frequency Identification) data, used to detect the
proximity.
Predic et al. [16] have developed a LBS and
mobile GIS application framework named
GinisMobile. It includes support for management
725
The First International Workshop on Managing Data with Mobile Devices
and presentation of raster and vector spatial data, as
well as dynamic data about mobile objects. The
contextual data used in GinisMobile is encoded
according to defined XML schema and transferred
to the server for analysis and storage. On the client
side, XML profile is parsed and used to customize
user interface.
Considering the ideas and solutions presented
in the works described above, we propose an
architecture for developing applications for context-
aware Mobile GIS, which is described in detail in
the next chapter. The architecture proposed in this
paper differs from other studies by proposing the
reuse of a domain-independent context model,
which is integrated to geo-spatial data management.
This feature allows the development of a Mobile
GIS capable of managing context information in
several areas of application.
5 CONCLUSIONS
This work presented the specification of an
architecture for the development of GIS-Mobile
systems capable of managing context information.
The characteristics inherent in this type of
application was studied in order to reach the
proposed solution, which includes: the use of Web
Services to allow remote access to information
from the mobile device, ontologies for context
modeling and use of geographic database to store
spatial data sets.
Using Web Services is an interesting solution
to solve problems of lack of computational
resources in mobile devices. This allows the
implementation of more complex functionalities in
more robust servers. Another advantage of using
Web Services is the interoperability between
different systems, which are described in different
languages. An example is the application UFV-
GeoMobile in which the system was implemented
in C# and the Web Service in Java.
The use of ontologies to specify a context
model has as the main advantage the possibility of
specifying the correct meaning and relationship
between the terms, avoiding ambiguous
interpretations of the domain being modeled.
Choosing the SeCoM model can facilitate modeling
a great deal, because it addresses the various
dimensions of contextual information and it was
developed with the characteristic of domain
independence.
A database diagram can be generated from an
ontology by adding the information of the type of
data and translating the formalism of the ontology-
based knowledge representation into a database
representation model as relational or object oriented
model. Similarly, a database diagram can be used to
establish the concepts to populate an ontology [17].
This approach to ontology-database mapping was
used in the development of the application UFV-
GeoMobile and was proven as a good way for
UbiCC Journal Volume 4
defining and obtaining context information.
The use of ontologies in this work could be
better exploited if there were a mechanism to
handle the ontologies and infer new information
before returning them to the Mobile GIS user. This
will be addressed in a future work.
Another feature that can be further exploited is
the use of Geo Web Services, which are services
defined by the OGC (Open GIS Consortium) to
address specifically geo-spatial data. This
mechanism would avoid the need for installing an
application on PDA, as it was done with the UFV-
GeoMobile system. Instead, the Mobile GIS appli-
cation could be accessed with a browser.
However, the architecture here presented,
together with the UFV-GeoMobile developed as
case study, show how the use of a Mobile GIS
application capable of managing information in
context can be very useful, especially the ability to
provide personalized information combined with
the display of geo-spatial data. Thus, this work
presents a very practical and feasible way to
develop applications for mobile GIS, presenting
solutions to address typical challenges of
applications involving geo-spatial data, context and
mobile devices.
ACKNOWLEDGMENTS
This project was partially supported with funds
from FAPEMIG - Foundation for the Support of
Research of the State of Minas Gerais and CNPq /
MCT / CT-Info.
6 REFERENCES
[1] Bulco Neto, R. F. and Pimentel, M. G. C.:
Toward a domain-independent semantic model
for context-aware computing. In Proceedings of
the 3rd Latin American Web Congress (LA-
Web05), pp. 6170, Buenos Aires, Argentina.
IEEE CS Press. (2005).
[2] Cad2Shp: Cad2Shp: AutoCAD DXF/DWG to
ArcView Shapefile Converter. Available at:
<http://www.brightergraphics.co.uk/guthrie/
cad2shp.htm>. Access: February 12, 2006
(2006).
[3] Dey, A. K.: Providing architectural support for
building context-aware applications. Estados
Unidos. PhD thesis (Doctor in Computer
Science) - Georgia Institute of Technology
(2000).
[4] Toivonen, S.; Kolari, J.; Laakko, T.: Facilitating
mobile users with contextualized content. In:
Procedings of Workshop Artificial Intelligence
in Mobile System (AIMS 2003).
[5] Li, X.; Shin, W.; Li, L.; Yoo, S. B.: GIS Web
Service Using Context Information in Mobile
Environments. In: Proceedings of ICCSA 2006,
LNCS 3980, pp. 895 903 (2006).
726
The First International Workshop on Managing Data with Mobile Devices

[6] Lisboa Filho, J.; Iochpe, C.: Modeling with a LNCS 3833, pp. 143-154 (2005).
UML profile. In: Shashi Shekhar and Hui [19] Castro, A.F.; Souto, M. V. S.; Amaro, V. E.;
Xiong. Encyclopedia of Geographical Vital H.: Desenvolvimento e Aplicao de um
Information Science. New York: Springer. pp. Banco de Dados Geogrficos na Elaborao de
691-700 (2008). Mapas da Morfodinmica Costeira e
[7] SharpMap: SharpMap: Geospatial Application Sensibilidade Ambiental ao Derramamento de
Framework for the CLR (2008). Available at: leo em reas Costeiras Localizadas no
<http://www.codeplex.com/SharpMap>. Estado do Rio Grande do Norte. Revista
Access: August 17, 2008. Brasileira de Geocincias. pp. 53-64 (2003).
[8] Solyman, A. A.: Investigating Mobile GIS [20] Arajo, M.A.: Web Services na Informao
(2005). Available at: <http://www.direction Geogrfica. Braga, Portugal. Dissertation
smag.com/article.php/>. Access: May 28, 2007. (Master in Informatic) Universidade do
[9] Strang, T.; Linnhoff-Popien, C.: A Context Minho, Portugal (2005).
Modeling Survey. In: Proceedings of Workshop
on Advanced Context Modelling, Reasoning
and Management (UbiComp4). Nottingham-
England (2004).
[10] Tomko, M.: Spatial Databases for Mobile GIS
Applications. Eslovquia. Dissertao
(Mestrado em Engenharia Civil) Slovak
University of Technology (2003).
[11] Weakliam, J., et al.: Delivering Personalized
Context-Aware Spatial Information to Mobile
Devices. In: Proceedings of W2GIS 2005,
LNCS 3833, pp. 194205 (2005).
[12] Weienberg, N.; Gartman, R.; Voisard, A.: An
Ontology-based Approach to Personalized
Situation-aware Mobile Service Supply.
GeoInformatica, v.10, n.1, pp. 55-90 (2006).
[13] Brisaboa, N. R.; Luaces, M. R.; Parama, J. R.;
Viqueira, J. R.: Managing a Geographic
Database from Mobile Devices Through OGC
Web Services. In: Procedings of
APWeb/WAIM 2007 Ws, LNCS 4537, pp.
174179 (2007).
[14] Sadeh, N. M.; Chan, E.; Van L.: MyCampus:
An agent-based environment for context-aware
mobile services. In: Procedings of Workshop
on Ubiquitous Agents on Embedded, Wearable
and Mobile Devices (UBIAGENTS 2002).
Bologna, Itlia (2002).
[15] Brovelli, M.; Magni, D.; Brioschi, M.; Legnani,
M.; Corcoglioniti, F.: NAMGIS - A Context-
Aware Mobile Web GIS. FOSS4G 2008, North
America. (2008).
[16] Predic B., D. Stojanovic and S. Djordjevic-
Kajan,:.Developing Context Aware Support in
Mobile GIS Framework.In: Proceedings 9th
AGILE International Conference on
Geographic Information Science, Visegrd,
Hungrary, pp. 90-97 (2006).
[17] Viegas, R. F.: GeOntoQuery: um mecanismo
de busca em banco de dados geogrficos
baseado em ontologias. Natal, RN, Brasil.
Dissertation (Master in Computer Science)
Universidade Federal do Rio Grande do Norte
(2006).
[18] Caduff, D.; Egenhofer, M. J.: Geo-Mobile
Queries: Sketch Based Queries in Mbile
Enviroments. In: Procedings of W2GIS 2005,

UbiCC Journal Volume 4 727