A Novel Hybrid Student Model for Personalized Education - Advances in Personalized Web-Based Education - page 76

Fig. 3.8 An example of the fuzzy weighted qualitative overlay model of the presented program-
ming tutoring system
knowledge, analytical thought) and strategic (integrated knowledge). De Raadt sug-
gests five knowledge levels:
• No answer : no knowledge
• Pre-structural : substantial lack of knowledge
• One-structural : ability to describe a part of code
• Multi-structural : ability to describe a program line-line
• Relational : ability to describe the whole of a program.
Therefore, the eight stereotypes of the KL stereotype category (Table 3.2 ) were
defined considering the above frameworks. A learner is classified to a KL stereotype
category according to which chapters the learner knows and how well. That kind of
information is derived from the fuzzy-weighted qualitative overlay model. The KL
stereotype category to which the learner has been classified, gives information about
the learning material that should be delivered to the learner (Fig. 3.9 ).
3.5.3.2 ErrTyp Stereotypes Category
ErrTyp stereotype takes two values: prone to syntax errors and prone to logical
errors. Syntax errors are recognized if they belong in one of the following
categories: anagrammatism of commands' names, omission of the definition of
data, using invalid command names etc. They, usually, indicate that the learner
has not read carefully and has not known adequately the chapters that correspond
to her /his knowledge level. Logical errors are usually errors of design and occur