Volume 19 Issue 9 - September 23, 2011 PDF
Counter
B2 Model: A Browsing Behavior Model Based on High-Level Petri Nets To Generate Behavioral Patterns for E-Learning
Yi-Chun Chang, Ying-Chia Huang and Chih-Ping Chu*
Department of Computer Science and Information Engineering, National Cheng Kung University
 
Font Enlarge
The Internet use has promptly proliferated in these years, causing e-learning systems to become increasingly popular. Developing an e-learning system with high quality, such as with intelligence and/or adaptive, has become a trend. In general, to verify or test the effect of an e-learning system, the researchers need to collect large amounts of learners’ browsing behavioral patterns. However, collecting sufficient behavioral patterns usually takes a great deal of time and effort. This paper has proposed the browsing behavioral model, B2 model (Fig. 1), to model and to generate students’ behavioral patterns. The experimental results have confirmed that the generated behavioral patterns based on the B2 model are similar to the browsing behaviors of actual students. In terms of practicability, the generated behavioral patterns based on the B2 model can serve as test data to validate the accuracy of an intelligent tutoring system. Therefore, the B2 model can facilitate the verification process of an intelligent tutoring system. In this purpose, since the B2 model is used to generate the behavioral patterns for system verification, the parameters which contain the average and variation values of the browsing time a student remains in a learning unit (BT), the time that a student undertakes an assessment (AT), and the score that a student is assigned for an assessment (AS), which are defined in the equations proposed in the paper, can be randomly set or be assigned according to the requirement of verification. For application, the B2 model is capable of predicting learning content that enables students to thrive in adaptive learning environments. For the purpose of prediction, the behavior of a few students needs to be collected to simply analyze the characteristics of students. The simply analyzed result can be used as a reference for which the average and variation values of BT, AT and AS defined in our proposed equations can be assigned. To verify the feasibility and practicability of the B2 model a tool was accompanied to develop in this research (Fig. 2).
Fig. 1

Fig. 2
< Previous
Next >
Copyright National Cheng Kung University