Bayesian and Fuzzy Logic Student Model in the C++ STL Intelligent Tutoring System

摘要

The C++ Standard Template Library (STL) Intelligent Tutoring System seeks to guide students in applying the STL for problem-solving. It is discovered that the key problem in using the C++ STL lies in the lack of capability in prerequisite concepts. Therefore, the ability to model a cause-effect relationship in Bayesian reasoning using a corresponding set of conditional probability is highly appropriate for the student modeling. To enhance the student model, a stereotype is assigned according to the student's understanding for further assessments. Fuzzy logic technique is capable of providing human-like diagnosis of the student's knowledge. The development applies practices from the eXtreme Programming methodology and J2EE technologies. The evaluation results revealed that the Bayesian Theorem has the capability of modeling the student's prerequisite and directing the student during the tutorial session. The Fuzzy Stereotyping of Students Expert System works well in categorizing the students according to four stereotypes - novice, beginner, intermediate and advanced.