Evaluation - Advances in Personalized Web-Based Education - page 114

“Sig. (2-tailed)” (Table 4.18 ) is lower than 0.05. Consequently, the system's
decisions about adaptive responses to the learner's needs are valid, since learners
seem to assimilate the learning material and succeed a better performance finally.
In addition, the percentage of times that a learner needed, finally, to read a
domain concept that the system had advised her/him not to read it is 7.14 %. This
percentage is sufficiently satisfactory to be able to lead to the conclusion that the
decisions, which are made by the system based on the student model, are valid.
4.4 Conclusions
The system's evaluation revealed that the combination of fuzzy sets with overlay
model and stereotypes contributes, significantly, to the adaptation of the learning
process to the learning pace of each individual learner. The results of the evalu-
ation demonstrated learning improvements and successful adaptation to students'
needs. In particular, the learners' overall rating of the presented web-based pro-
gramming tutoring system is very high. The participant learners were very
satisfied with the quality of content and quality, with the tutoring system's friend-
liness and usefulness, and with the system's adaptive responses to their needs.
Furthermore, the integration of the presented novel fuzzy student model into the
programming tutoring system improved significantly the student's performance.
Also, the learners, who used the presented e-learning tutoring system, obtained
a more positive state and behavior towards computer programming and distance
learning. The assessment results showed, also, that the e-learning program helped
learners to their further studies satisfactory. In addition, the evaluation results
revealed that the presented novel fuzzy approach of student modeling improves
the efficiency of the adaptation of the instructional process. In particular, both,
the conclusions that are drawn by the system concerning the aspects of students'
characteristics and the adaptation decision-making were valid. The system advises
the student model and adapts instantly the sequence of learning lessons concerning
the students' characteristics with success. Consequently, the presented novel hybrid
student model contributes significantly to the adaptation process and helps the
system to provide a personalized and effective educational process for learning of
computer programming.