Student Modeling for Personalized Education: A Review of the Literature - Advances in Personalized Web-Based Education - page 12

incorporated an overlay student model to represent the student's knowledge level
in order to provide effective and personalized selection of the appropriate learning
resources.
Another student modeling technique that is usually used to model the learner's
knowledge level is stereotyping. Examples of adaptive and/or personalized tutoring
systems that have used stereotypes for modeling the student's knowledge lever are the
following. AUTO-COLLEAGUE (Tourtoglou and Virvou 2008, 2012), which is an
adaptive and collaborative learning environment for UML, represents the level of
students' expertise through a stereotype-based modeling technique. Furthermore,
Chrysafiadi and Virvou (2008) have developed a stereotyping approach to model
the knowledge level of learners in the programming language Pascal in order to
adapt the system's responses to each individual student dynamically. Also, a stere-
otype-like approach for modeling the student's knowledge level is used in Wayang
Outpost, which is a software tutor that helps students learn to solve standardized-
test type of questions, in particular for a math test called Scholastic Aptitude Test,
and other state-based exams taken at the end of high school in the USA, in order to
discern factors that affect student behavior beyond cognition (Arroyo et al. 2010).
Moreover, Durrani and Durrani (2010) have used stereotypes for modeling the
student's knowledge the adaptive C ++ tutor CLT. Finally, Grubiši et al. (2013)

have defined knowledge stereotypes based to model the student's proficiency in
an adaptive e-learning system called Adaptive Courseware Tutor (AC-ware Tutor).
Another technique of modeling the learner's knowledge is the Constraint-
Based Model (CBM). Mitrovic (2003) have used the CBM approach to model
the student's knowledge of a web-enabled intelligent tutoring system that teaches
the SQL database language. Another system that uses CMB for modeling the
student's knowledge is COLLECT-UML, which is an ITS that teaches object-
oriented design using Unified Modeling Language (Baghaei et al. 2005). Also,
Weerasinghe and Mitrovic (2011) have applied CBM to model the student's
knowledge in EER-Tutor, which is an ITS that teaches conceptual database design.
Furthermore, BNs have been used for the representation of the student's
knowledge. For example, Bunt and Conati (2003) used Bayesian Networks to
detect when the learner is having difficulties in an intelligent exploratory learning
environment for the domain of mathematic functions. A Bayesian student model
was applied in English ABLE for modeling the student's knowledge in English
grammar (Zapata-Rivera 2007). Furthermore, in TELEOS a Bayesian network
based student model was used in order to explicitly diagnose the student's knowl-
edge level (Chieu et al. 2010). Similarly, AdaptErrEx has used BNs to model
learners' skills (Goguadze et al. 2011a, b). Also, INQPRO system predicts the
acquisition of scientific inquiry skills by modeling students' characteristics with
Bayesian networks (Ting and Phon-Amnuaisuk 2012).
Several student models for learners' knowledge representation have been built
based on ontologies. For example, MAEVIF (Clemente et al. 2011) and SoNITS
(Nguyen et al. 2011) have used ontologies to model the student's knowledge.
Also, Peña and Sossa (2010) have adopted a semantic representation and manage-
ment of student models with ontologies in order to represent learners' knowledge.

ć