Test Paper Analysis Based On Bayesian Network

Test is a necessary component of the whole teaching course. It serves as a check on the effect of both the teaching and the learning. The investigation on the result of the test will help us reflect on the teaching course, and find what to do next and how to make improvements.The Bayesian Network (BN) proposed by Pearl is a new mechanism for uncertain knowledge representation and manipulation based on probability theory and graph theory. BN is network structure with clarity semantics. It exploits the structure of the domain to allow a compact representation of complex joint probability distribution. Its sound probabilistic semantics explicit encoding of relevance relationships, inference algorithms and learning algorithms that are fairly efficient and effective in practice, and decision-making mechanism of facility, have led BN to enter the Artificial Intelligence(AI) mainstream.The present thesis is to make an experimental analysis of the test paper based on Bayesian Network. The main toolkit used in this experiment is BNT software suite compiled with MATLAB. This software suite provides us with a lot of basic function sets for Bayes Network learning. It is suitable for the accurate and appropriate logics of various types of joints, and it also has the function of parameter learning and structure learning. From the experiment we come to the conclusion that five factors including student-attendance and fulfillment-of-homework have great influence on their scores.