Supporting E-Learning System with Modified Bayesian Rough Set Model

摘要

The increasing development of Internet, especially Web-based learning is one of the most important issues. In this paper, a new application on Bayesian Rough Set (BRS) model for give information about learner performance is formulated. To enhance the precision of original rough set and to deal with both two decision classes and multi decision classes, we modify BRS model based on Bayesian Confirmation Measures (BCM). The experimental results are compared with that got by other methods. The quality of the proposed BRS model can be evaluated using discriminant index of decision making, which is suitable for providing appropriate decision rules to the learners with high discriminant index.