A Study Of Concept Map Based Learning Path Control For Web Learning

Online learning is the focus of current discussion. Increasing courses are delivered through Web, and the Web has become an important means to support distance instruction and lifelong learning. However, some research found that learners often feel disoriented and information overloaded in hypermedia space. While disoriented, the learner doesn't know where to go, or knows where to go but doesn't know how to get there, or even doesn't know where he is. Disorientation leads to information overloading, and vice versa. Such a vicious circle may affect learner's confidence in online learning.According to the trends of online learning and the two crucial problems in hypermedia space, this research supposed that the existing online learning systems or web courses have a lot to change on offering students appropriate guidance. And the most recent educational researches on concept map not only made it a way of visualized knowledge representation, but its ideas, such as node, link, cross link and hierarchy, can also be used in the design and development of adaptive hypermedia learning systems. Thus the research suggested a "Cmap-based Learning Path Control Mechanism", which can identify the misconceptions of an individual learner through diagnostic and formative evaluation strategies, and generate a system-controlled guiding path according to individual learner's knowledge level. The mechanism enhances the adaptivity of web hypermedia learning system, and better satisfies the needs of learners.In order to validate the feasibility of this method, the reseach group developed a prototype system, using the first lesson of multimedia technology course. The work can be summed as follows: First, decompose and reconstruct adaptive web course structure through the structural knowledge representation of concept map. Second, identify individual learner's misconceptions through item attributes table and the improved concept cumulative-count algorithm. Third, generate a system-controlled guiding path according to the learner's misconceptions, and lead the learner going across knowledge area to learning goals. Finally, let five types of experts and a small group of students try out the system, and conduct the formative evaluation of CLPCS prototype.Through the analysis of research problem, the design, development, utilization and evaluationof CLPCS system, the research concluded: The structural knowledge representation of concept map can effectively decompose and construct adaptive web course structure, and relate quiz items to content modules. Learner model based on domain model and concept hierarchy can describe an individual learner's knowledge level, it is the basis of system's adaptive behaviors. Carefully designed unit quizzes can improve learner-system interaction, identify learner misconceptions, and maintain learner's learning interest in web environment. The individualized, system-controlled path can become a helpful guide when students meet with learning obstacles, and can reduce the possibility of reading redundant information. The learner's control over the system should always be considered preferably, for system's adaptivity and intelligence is quite limited while comparing with human being's. During the formative evaluation procedure, experts and students told their subjective feelings of this system's instructional effectiveness and gave some advice on how to improve it. The advice can be used for further work. The theory model and prototype system still need to be validated, improved, extended and refined in future.