LOCATION BASED USER MODELING IN ADAPTIVE MOBILE

LEARNING FOR ENVIRONMENTAL AWARENESS

Efthimios Alepis

, Maria Virvou

and Katerina Kabassi

Department of Informatics, University of Piraeus, 80 Karaoli & Dimitriou St., 18534, Piraeus, Greece

Department of Ecology and the Environment, Technological Educational Institute of the Ionian Islands

Zakynthos, Greece

Keywords: Mobile learning, Environmental informatics, User modelling, Adaptive learning.

Abstract: Recently it has been widely acknowledged that the incorporation of advanced information technologies in

the areas of ecological informatics may provide significant assets towards social environmental awareness.

In this paper we present a sophisticated mobile learning system which offers environmental educational

information to users based on their current geographical location. The system adapts its content according to

the user’s personal characteristics and to the user’s mobile device. The adaptation of the user’s interface is

accomplished through the incorporation of a well known decision making model, namely the Analytic

Hierarchy Process (AHP). The resulting prototype system is called m-AWARE and is targeted to people of

all ages, providing easily accessible information about our environment for environmental awareness

purposes.

1 INTRODUCTION

Environmental informatics is the application of

information technology to environmental science. In

(Rickinson, Lundholm and Hopwood, 2010), it is

stated that the last 4 decades have seen growing

international recognition for the educational

dimensions of environmental and sustainable

development issues. Since the late 1960s,

international statements from organizations such as

the IUCN (International Union for the Conservation

of Nature) (www.iucn.org) and UNESCO (United

Nations Educational, Scientific and Cultural

Organization) (www.unesco.org) have called by

environmental problems to be tackled through

environmental education for all age groups. The

need for prompt and valid information is even

greater nowadays, since we are all witnesses of daily

environmental disasters and of the irreversible

damages to our ecosystem. Environmental disasters

can have an effect on agriculture, biodiversity,

economy and human health. The causes include

pollution, depletion of natural resources, industrial

activity or agriculture.

A remedy against the rapid destruction of our

natural environment due to human activities may lie

at providing easily accessible and comprehensive

environmental data to all the people through the use

of recent technological and scientific achievements.

As it is stated in (Pillmann, Geiger and Voigt, 2006),

in the science sector, a rapidly growing community

conceived new computer applications for decision

making and information exchange in the field of

environmental protection, since environmental

problems in the last decades have resulted in an

increased ecological awareness in Europe.

Ecological informatics (ecoinformatics) is an

interdisciplinary framework for the processing,

archival, analysis and synthesis of ecological data by

advanced computational technology (Recknagel,

2006). According to (Recknagel, 2006),

computational technologies currently considered

being crucial for data archival, retrieval and

visualization in Environmental Informatics include:

 Object-oriented data representation to facilitate

data standardization and data integration by the

embodiment of metadata and data operations into

data structures;

 Internet and world wide web to facilitate

interactive and online simulation as well as software

and model sharing;

 Remote sensing and GIS to facilitate spatial data

visualization and acquisition;

214

Alepis E., Virvou M. and Kabassi K..

LOCATION BASED USER MODELING IN ADAPTIVE MOBILE LEARNING FOR ENVIRONMENTAL AWARENESS.

DOI: 10.5220/0003505302140217

In Proceedings of the 6th International Conference on Software and Database Technologies (ICSOFT-2011), pages 214-217

ISBN: 978-989-8425-76-8

 2011 SCITEPRESS (Science and Technology Publications, Lda.)

 Adaptive agents to facilitate adaptive simulation

and prediction of ecosystem composition and

evolution.

In view of the above, in this paper we present a

novel adaptive mobile learning system that

incorporates all the four fore mentioned computation

technologies. The system’s main objective is to use

location based and user modeling information for

environmental awareness purposes. The resulting

system is called m-AWARE, which is the acronym

for “mobile-Adaptive Warnings and Advice for

Resources of the Environment”. More specifically,

in this paper we focus on the application of recent

advances in Information Technology, such as mobile

software engineering, multi-criteria decision

making, adaptive hypermedia and geographic

information systems (GIS) to environmental science.

The proposed theory for the construction of the

multi criteria decision making model is the Analytic

Hierarchy Process (AHP) method. The AHP method

is used as a reasoning mechanism for the

specification of the information that is delivered to

the users through their mobile devices. Each user’s

profile includes information about the specific user

(such as user’s age, user’s educational background,

interests, gender, etc.), as well as information about

the users’ current geographic location. Finally,

information about each user’s personal mobile

device will be also retrieved in order to adapt the

application to the user’s device needs and

limitations. The proposed system will also use

stereotypic information derived from each user’s

given personal information.

The architecture that is used for the

representation of the available data is based on the

Object Oriented model. Object oriented approaches

have been already widely used in software

development environments (Chiu, Lo and Chao,

2009), (Pastor, Gomez, Insfran and Pelechano,

2001). The resulting system is able to process

ecological data and present the appropriate

information to users who own mobile devices based

on their personal profile, where they are (geographic

location in a specified range), and what mobile

device they are using. Accordingly, the

representation of the available ecological

educational information is dynamically adapted to

each user. Finally, the interaction between users and

the application is friendly to a high extent through

the use of pedagogical animated agents.

2 DECISION MAKING MODEL

THROUGH THE ANALYTIC

HIERARCHY PROCESS

AHP is one of the most popular Multi Criteria

Decision Making (MCDM) methods. It has solid

theoretical foundation and objectivity to some

degree. AHP is based on three principles:

decomposition, comparative judgments, and the

synthesis of priorities, and can help decision makers

to develop systematic approaches for a variety of

problems.

The Analytic Hierarchy Process (AHP) (Saaty,

1980) is composed of several previously existing but

unassociated concepts and techniques, such as

hierarchical structuring of complexity, pair wise

comparisons, an eigenvector method for deriving

weights etc. (Jandric and Srdjevic, 2000), (Selly and

Forman, 2001). Based on mathematics and

psychology, the AHP has been extensively studied

and refined over the last decades. It provides a

comprehensive and rational framework for

structuring a decision problem, for representing and

quantifying its elements, for relating those elements

to overall goals, and for evaluating alternative

solutions.

The method consists of the following steps (Zhu

and Buchman, 2000):

 Developing a goal hierarchy.

 Setting up a pair wise comparison matrix of

criteria.

 Ranking the relative importance between

alternatives.

 Checking consistency of the comparisons.

Calculating AHP values.

The AHP value is computed using the following

formula:





jiji

waAHP

, for

Mi ,...,3,2,1

where M is the number of alternatives and N is the

number of criteria; a_{ij} denotes the score of the

i^th alternative related to the j^{th} criterion; W_j

denotes the weight of the J^{th} criterion.

Figure 1, illustrates the AHP hierarchy that

results by the application of the AHP’s model to our

system.

The exact weights for the criteria that are used in

our implementation of the AHP model have been

initially specified by the authors. However, a future

empirical study may reveal more accurate values for

the determination of each criterion’s importance.

LOCATION BASED USER MODELING IN ADAPTIVE MOBILE LEARNING FOR ENVIRONMENTAL

AWARENESS

215

Figure 1: AHP hierarchy in m-AWARE.

3 OVERVIEW OF THE SYSTEM

Figure 2: Architecture of m-AWARE.

In this section, we describe the overall functionality

and features of m-AWARE.

The architecture of m-AWARE consists of the

main educational application, a user modeling

component, a decision making inference mechanism

and a database. Part of the database is used to store

educational data and another part is used to store

data related to user modeling. Accordingly, the

database is also used to store user models and user

personal profiles for each individual user that uses

and interacts with the system, as well as stereotypic

information about user profiles. The system’s

architecture is illustrated in figure 2.

As we can see in figure 2, the students’

interaction can be accomplished either orally

through the mobile device’s microphone, or through

the mobile device’s keyboard. The educational

system consists of three subsystems, namely the user

modeling subsystem, the educating application

subsystem and the subsystem that incorporates the

decision making mechanism. Both the user

modelling subsystem’s data and the educational

subsystem’s data are stored in the main system’s

database, while the decision making subsystem is

responsible for the resulting interface created for

each user during his/her interaction with m-

AWARE. m-AWARE has been developed to operate

on the Android mobile operating system, while as

for future work the authors are planning to provide

implementations for other existing mobile phone

platforms as well. Correspondingly, the system is

programmed using JAVA as a programming

language. This specific programming language is

also compatible with the system’s Object Oriented

structure. Figure 3 illustrates a snapshot of the

operating educational application, where a user is

retrieving environmental information about a

specific geographic location.

ICSOFT 2011 - 6th International Conference on Software and Data Technologies

216

Figure 3: A user is viewing the available information

adapted to his/her profile.

4 CONCLUSIONS AND FUTURE

WORK

Greece, has the tenth longest coastline in the world

at 14,880 km in length, featuring a vast number of

islands (approximately 1400, of which 227 are

inhabited). Eighty percent of Greece consists of

mountains or hills, making the country one of the

most mountainous in Europe. However, continuous

environmental disasters have negative effects on our

country’s agriculture, environment and tourism and

as a result to our economy and human health. A

remedy to such environmental problems may lie in

providing environmental education for all age

groups. In our research, we have aimed in creating a

novel Adaptive Mobile Learning system for

Environmental Awareness. It is in our future plans

to evaluate m-AWARE in order to examine the

degree of its usefulness as an educational tool, as

well as the degree of usefulness and user-

friendliness for the people who are going to use the

educational system.

REFERENCES

Chiu, P.-H., Lo, C.-C. , Chao, K.-M., 2009. Integrating

semantic web and object-oriented pro-gramming for

cooperative design, Journal of Universal Computer

Science, Volume 15, Is-sue 9, Pages 1970-1990

http://www.iucn.org/

http://www.unesco.org/

Jandric, Z. and Srdjevic, B., 2000. Analytic Hierarchy

Process in Selecting Best Groundwater Pond, 31st

International Geological Congress

Pastor, O., Gómez, J., Insfrán, E., Pelechano, V., 2001.

The OO-Method approach for information systems

modeling: From object-oriented conceptual modeling

to automated programming, Information Systems, Vol

26, Issue 7, pp. 507-534

Pillmann, W., Geiger, W. and Voigt, K., 2006. Survey of

environmental informatics in Europe, Environmental

Modelling and Software, Volume 21, Issue 11, Pages

1519-1527.

Recknagel, F., 2006. Ecological Informatics, Springer

Rickinson, M., Lundholm, C., Hopwood, N., 2010.

Environmental Learning: Insights from research into

the student experience, Springer.

Saaty, T., 1980. The Analytic Hierarchy Process:

Planning, Priority Setting, Resource Allocation, ISBN

0-07-054371-2, McGraw-Hill.

Selly, M. S. and Forman, E. H., 2001. Decision by

Objectives, World Scientific Publishing Co.

Zhu, Υ., Buchman, Α., 2000. Evaluating and Selecting

Web Sources as External Information Resources of a

Data Warehouse, The Third International Conference

on Web Information Systems Engineering (WISE'00),

149-160.

LOCATION BASED USER MODELING IN ADAPTIVE MOBILE LEARNING FOR ENVIRONMENTAL

AWARENESS

217