Computerized Evaluation and Diagnosis of Student's Knowledge Based on Bayesian Networks

In this paper, we describe the integration and evaluation of an existing generic Bayesian student model into an existing computerized testing system within the Projecto Matematica Ensino (PmatE) of the University of Aveiro. The Bayesian student model had previously been evaluated with simulated students, but a real application was still needed. The testing system in PmatE is based in the use of Learning Objects (LO), which are question generators which essentially consist of some parameterized text and sets of parameterized "true/false" questions (at least four). These LO together with the experience of PmatE in using computerized tests with students, gives us ideal conditions for testing the described Bayesian student model with real students.