Information Technology Reference
In-Depth Information
AWARD
prime
: An Adaptive Web Based-Tool
Prototype for Neurocognitive Individualized
Assessment and Training
Raquel Salmeron
1
,Seraf ın Crespo
1
,FranciscoLopez
2
,
Mar ıa Teresa Daza
2
, and Francisco Guil
1
1
Dept. de Lenguajes y Computacion - Universidad de Almerıa
{
rsalmeron,scp122,fguil
}
@ual.es
2
Dept. Neurociencia y Ciencias de la Salud - Univeridad de Almerıa
{
fll416,tdaza
}
@ual.es
Abstract.
As the web is expanding and the underlying technologies are
evolving, so does the way in which users interact with the web applica-
tions. It is becoming increasingly clear that web applications must be
adapted to the specific characteristics of each user, resulting in what is
known as Adaptive Web Applications. An obvious fact is that there is no
a single user model. Actually, there are many factors, some quantifiable,
which define how a user interacts with a particular web application, al-
lowing the determination of the user profile and setting the associated
default values of each adaptive variables. This paper presents a proto-
type tool whose core consists of an adaptive method based on similarity
between (mined) event-based sequences. The application domain belongs
to the area of cognitive neuroscience. In particular, the tool has been de-
signed as a platform for the specification of adaptive cognitive tasks,
created to assess and stimulate different aspects of cognition in patients
with neurological damage.
1
Introduction
Learning and evolution are two fundamental forms of adaptation. The adap-
tive web emerges as a solution to satisfy, depending on the interaction with the
web application, the particular necessities of each specific user [9]. An adap-
tive web application is defined as an application that automatically improve its
organization and presentation by learning from users access patterns.
Usually, adaptive web applications mine the data buried in web server logs to
produce more easily navigable web applications. We find in the literature several
approaches for adapting the web structure depending of the particularities of
each user. These approaches can be divided into two groups: explicit and implicit
methods. The explicit methods determine user profiles based on an enumeration
of predetermined rules. On the other hand, the implicit methods extract the user
profiles from a set of discovered patterns from web server logs [8]. Following the
same logic as the implicit methods, we present an adaptive web-based application
Search WWH ::
Custom Search