Student Modeling for Personalized Education: A Review of the Literature - Advances in Personalized Web-Based Education - page 10

1.2.7 Ontology-Based Student Modeling
Recently a lot of research has been done on the crossroad of user modeling and
web ontologies. Due to the fact that the adaptive and/or personalized tutoring sys-
tems attempt to model the teaching and learning processes in real world and the
most of them are web-based applications, they can be combined with web ontolo-
gies. Ontologies support the representation of abstract enough concepts and prop-
erties and make them reused and extended in different application (Clemente et al.
2011). These characteristics of ontologies can help student modeling. The main
advantages of ontology-based student models are: formal semantics, easy reuse,
easy probability, availability of effective design tools, and automatic serialization
into a format compatible with popular logical inference engines (Winter et al.
2005).
1.3 Student's Characteristics to Model
A significant initial stage of constructing a student model is the selection of appro-
priate students' characteristics that should be considered and represented. The per-
sonalization is accomplished efficiently by modeling either the domain dependent
student's characteristics or the domain independent domain student's character-
istics (Yang et al. 2010). For example, domain dependent student's characteris-
tics are the knowledge level, the misconceptions, and the prior knowledge. Some
domain independent student's characteristics are learning style, memory, concen-
tration, and self-assessment. The student's characteristics are, also, categorized
into static characteristics (like email, age, native language) and dynamic character-
istics (like knowledge level, errors). The static characteristics are set by the student
at the beginning of the learning process, usually through questionnaires, while the
dynamic characteristics are defined and updated each time the student interacts
with the system.
Therefore, the challenge is to define the dynamic student's characteristics that
constitute the base for the system's adaptation to each individual student's needs.
These characteristics include knowledge and skills, errors and misconceptions,
learning styles and preferences, affective and cognitive factors, meta-cognitive
factors (Fig. 1.9 ).
The student's characteristics that are usually modeled are:
1. Knowledge level
2. Errors and misconceptions
3. Cognitive features other than knowledge level
4. Affective features
5. Meta-cognitive features
These students' characteristics are described in detail in the following subsections.