Student Modeling for Personalized Education: A Review of the Literature - Advances in Personalized Web-Based Education - page 15

technique perturbation to model the student's errors and misconceptions. In par-
ticular, Surjono and Maltby (2003) have used a perturbation student model to per-
form a better remediation of student mistakes. Furthermore, LeCo-EAD (Faraco
et al. 2004) and InfoMap (Lu et al. 2005) have modeled students' misconceptions
by using a perturbation model. InfoMap's perturbation student model involves 31
types of addition errors and 51 types of subtraction errors (Lu et al. 2005). The stu-
dent model of both systems allows the reasoning of students' errors and helps the
system to expand the explanation during the feedback to the students. Moreover,
Baschera and Gross (2010) have represented through the perturbation approach
the student's strength and weaknesses, in order to allow for appropriate remedia-
tion actions to adapt to students' needs. A perturbation student model for detecting
the student's errors has been used in AUTO-COLLEGE (Tourtoglou and Virvou
2012).
Furthermore, there are adaptive tutoring systems that have used other techniques
than perturbation to model the student's errors and misconceptions. In particular,
Virvou and Kabassi (2002) have added more “human” reasoning to F-SMILE by
using stereotypes and cognitive theory of Human Plausible Reasoning (HPR)
(Collins and Michalski 1989). F-SMILE reacts accordingly trying to find out the
cause of the problematic situation in which the user is involved when s/he learns
how to manipulate file store of her/his computer. Goel et al. (2012) used a fuzzy
model for student reasoning based on imprecise information coming from the stu-
dent-computer interaction and performed the prediction of the degree of error a stu-
dent makes in the next attempt to a problem. Also, Chrysafiadi and Virvou (2008)
have modeled the type of programming errors that a student can make during
her/his interaction with a web-based educational application that teaches the pro-
gramming language Pascal (Web_Tutor_Pas) using stereotypes. Furthermore, so
KERMIT (Suraweera and Mitrovic 2004) that teaches conceptual database design,
as J-LATTE (Holland et al. 2009) and INCOM (Le and Menzel 2009), which teach
language programming, use the CBM approach to diagnose the student's errors. In
addition, AdaptErrEx has used BNs to model learners' misconceptions (Goguadze
et al. 2011a, b). BNs have been, also, used for modeling student's errors in Andes
(Shapiro 2005). Moreover, Pérez-de-la-cruz (2002) has modeled the student's
misconceptions applying BNs in combination with cognitive theories.
Therefore, there are a variety of student modeling techniques that can be
used to model the learner's errors and misconceptions. In Table 1.2 , the percent-
ages of preferences for each one of the student modeling techniques for mod-
eling the student's errors and misconceptions are presented considering the above
literature review. The information that is derived from the particular table is the
number of the adaptive educational systems that incorporate a particular student
modeling technique for the modeling of the learner's errors and misconceptions
in a set of one hundred adaptive educational systems. For example, if we have
a hundred adaptive educational systems 35.71 of them will use the perturbation
approach, 21.43 will use the Constraint-based model, 21.43 % will use Bayesian
Networks etc.