|DC.description||National Central University||en_US|
|dc.description.abstract||Many computer-assisted learning systems adopt immediate feedback to help students learning. However, the problem has been occurred that students using feedback inappropriately. Some students got passed to the next stage by constantly requesting for system feedback instead of actually learning the studying concepts which lead to fewer learn the content knowledge on the mathematics. In this study, we reference by mastery learning theory and adaptive learning to set a threshold of mastery. The threshold of mastery can give each student a dynamic number of exercises. So, students who have completed basic exercises and passed the threshold of mastery can go to the next stage without additional exercises. Otherwise, students who have not passed the threshold of mastery, the system will give them strengthened exercises to improve students′ mastery levels and their understanding of mathematical concepts.
This study was conducted in the ＂Math Island＂ system. In the first phase of the experiment, students were divided into two groups, the fixed-practice group (control group) and the dynamic-practice group (experimental group). The proficiency prediction model was established by utilizing the historical data of feedback level and learning effectiveness evaluation used by the students in the control group. According to the prediction model, a dynamic exercise mechanism was established and the number of exercises given to the dynamic-exercise group by dynamic adjust. This study hoped that the students will be able to master the learning content after completing the exercises by dynamic exercise mechanism. Moreover, this study also established different prediction models to find an appropriate prediction model that can accurately predict the evaluation of learning effectiveness.
The results show that the revised prediction model is more accurate than the non-revised one, and the proficiency prediction model using the sequence of feedback levels and the total number of feedback levels can accurately predict the evaluation of learning effectiveness. As far as the total number of exercises is concerned, the performance of the students in the dynamic-practice group is close to the fixed-practice group. It is found that the dynamic exercise mechanism can accurately predict the evaluation of learning effectiveness.
|DC.subject||Dynamic Exercise mechanism||en_US|
|DC.title||Implement a Dynamic Exercise Mechanism Based on Learning Effectiveness Prediction Model||en_US|
|DC.publisher||National Central University||en_US|