Dynamic Data Feed to Bayesian Network Model and SMILE Web Application

A complete diagnostic Bayesian network model cannot be achieved and the result of the constructed model cannot be guaranteed unless correct and reliable data are provided to the model. In this paper, we propose a technique to dynamically feed data into a diagnostic Bayesian network model in the first part of this paper. In the second part of the paper, a case study of several factors that have an impact on students for making a decision in enrollment is transformed into a Bayesian network model. The last part of the paper discusses a Web user interface for the model in terms of its design and diagnosis. The user is allowed to perform a diagnosis of the model through the SMILE Web application interface.