A Novel Hybrid Student Model for Personalized Education - Advances in Personalized Web-Based Education - page 70

knowledge on the domain concept Ci, i , or the type of errors (ErrTyp) that s/he
made. Then, the system applies the fuzzy rules in order to identify and update
the student's knowledge level of all the related concepts with Ci. i . The values
of the knowledge level of all the domain concepts of the learning material
that came off the application of the fuzzy rules are used for the definition of
the fuzzy-weighted qualitative overlay model. So, the overlay student model
is updated.
The new information of the overlay model triggers the KL stereotype
category. Therefore, the system decides to which stereotype category of knowl-
edge level, it should classify the learner concerning the learner's knowledge
level of the domain concepts of the learning material. If s/he succeeds in the
test, then s/he is transited to a next knowledge level and the value of KL is
increased. Otherwise, if s/he fails, then s/he remains to the same knowledge
level or s/he is returned to a previous knowledge level, according to her/his
errors. In particular, if the learner's poor performance does not affect the knowl-
edge level of other related domain concepts, which belong to a previous section,
then the value of KL remains the same; otherwise the value of KL is decreased.
If the learner makes errors that correspond to concepts of previous section,
then the system infers that s/he has forgot something from previous sections of
the learning material. In particular, if the learner makes errors, which consider
concepts of previous knowledge level, then the system checks the value of the
stereotype that corresponds to her/his prior knowledge on related fields. If the
system decides that the errors were made due to confusion with prior knowl-
edge, then it does not classify the learner to a previous knowledge level, but it
points out the error. Otherwise, it classifies the learner to a previous knowledge
level reducing the value of KL.
Therefore, the operation of F.O.S. is described in the following steps:
1. The learner registers into the system and enters her/his age, name and her/his
prior knowledge (PrK) on related fields.
2. The learning material is delivered to her/him.
3. S/he is examined in domain concept Ci i taking a test.
4. The system identifies the type of errors that the learner made (ErrTyp).
5. If the learner makes errors due to confusion with prior knowledge, then it does
not consider them into calculating her/his performance, but it points out the error.
6. The system identifies the learner's knowledge level on the concept Ci. i .
7. The system applies the fuzzy rules and defined the alterations on the learner's
knowledge level of all the related with Ci i concepts, considering either the
knowledge dependencies that exist between the domain concepts of the learn-
ing material or the current learner's knowledge level on the domain concepts of
the learning material.
8. The system updated the fuzzy-weighted qualitative overlay model.
9. The system advises the overlay model and classifies the learner to the appropriate
stereotype category of knowledge level (KL).