An Innovative Approach to Develop an Intelligent Virtual University

Virtual University is widely used as a means of communication between teachers and students. However its educational impact has always been under an aura of ambiguity. In contrast, Intelligent Tutoring System increases education quality by providing feedback and hint to students, but its implementation cost is high. Assessment of student knowledge and choosing an appropriate hint are important components of an Intelligent Tutoring System which have a significant role in the educational impact of the system. This study presents a new approach to implement these two components of intelligence at low-cost in a virtual university with a desired knowledge domain. A Bayesian Network and an Artificial Neural Network based on training data, designed for knowledge assessment and choosing appropriate hints respectively. This approach implemented and evaluated at the Virtual University of Sistan and Baluchestan in domain of search techniques in Artificial Intelligence. Accuracy of the Bayesian Network is evaluated to be 98% on average and accuracy of the Artificial Neural Network is evaluated to be 92.5%. Using Rules for knowledge representation and designing networks based on training data has given a good expandability to the proposed approach

کلیدواژه ها:

Intelligent Tutoring System ، Virtual University ، Bayesian Network ، Artificial Neural Network