A Novel Hybrid Student Model for Personalized Education - Advances in Personalized Web-Based Education - page 86

Consequently, the system infers that Stella's knowledge level has increased in more
concepts than C 4.2 . Thereby, it advises her overlay model and increases the value of
KL stereotype, classifying her to an upper knowledge level to read the following learn-
ing material. In particular, the value of KL becomes 5 declaring that the system consid-
ers that Stella does not need to read concepts that correspond to knowledge level 4.
Example 4 The parameters of Dimitris' student model had the following values:
KL = 6, ErrTyp = “prone to syntax errors”, PrK = “Pascal”. After having com-
pleted a test which involved exercises on two-dimensional arrays, the system dis-
covered that he made more than 40 % errors on the assignment statement, and more
specifically he used the symbol : = rather than the symbol = . The system checked
the values of the parameter PrK and was informed that he already knows the pro-
gramming language Pascal. Thus, it was assumed that Dimitris used the symbol
: = for assignment due to his previous knowledge on Pascal. So, the system' reac-
tion was to stress the error, but it did not classify him to the knowledge level 1.
Example 5 Alexis's current student model has the following values: KL = 4,
ErrTyp = “prone to logical errors”, PrK = “Pascal”. The value KL = 4 comes off
his current overlay model (Table 3.6 , column 'before'). ErrTyp is “prone to logical
errors” due to the fact that he had made usually errors that concern the seman-
tics and operation of the commands. PrK = “none” indicates that Alexis knows the
programming language 'Pascal'.
He is examining in C 5.4 : “calculating average in while 'for' loop” and is
succeeding 76 %. So, the quintet, which describes his knowledge level on C 5.4 ,
is (0, 0, 0.8, 0.2, 0). However, according to the “strength of impact” of the
knowledge dependencies that exist between the domain concepts of the learning
material (Table 2.2 ) , C 5.4 affects the concept C 4.2 , C 4.3 , C 4.4 , C 5.2 and C 5.3 .
According to the fuzzy rules (Figs. 3.6 and 3.7 ) the knowledge level of the
related concepts is deteriorated as follows (Table 3.6 , column 'after'):
• According to R7 C 4.2 remains 100 % 'Assimilated'.
• According to R8 is X 4.3 = ( 1 − 0.41 ) × 88.6 + MIN[0.41 × 88.6, 0.41 × 76] =
83.46. Therefore, KL(C 4.3 ) = L with μ L (C 4.3 ) = 1. So, the quintet for C 4.3 is
(0, 0, 0, 1, 0).
• According to R8 is X 4.4 = ( 1 − 0.52 ) × 89 + MIN[0.52 × 89, 0.52 × 76] = 84.8.
Therefore, KL(C 4.4 ) = L with μ L (C 4.4 ) = 1. So, the quintet for C 4.3 is (0, 0, 0, 1, 0).
• According to R7 C 5.2 remains 100 % 'Assimilated'.
• According to R8 is X 5.3 = ( 1 − 0.41 ) × 88.6 + MIN[0.41 × 88.6, 0.41 × 76] =
83.44. Therefore, KL(C 5.3 ) = L with μ L (C 5.3 ) = 1. So, the quintet for C 4.3 is
(0, 0, 0, 1, 0).
Thereby, the system infers that Alexis has forgotten the concepts C 4.3 , C 4.4 , C 5.2
and C 5.3 and it classifies him to the previous corresponding knowledge level to
revise them, reducing the value of KL to 3.
Example 6 George, a novice student, interacted with the system for the first time.
He completed the test that involved exercises, which concerned the basic concepts
of “C”. The results are depicted in Fig. 3.11 . Particularly, he does 29 % errors