The research is to implement an intelligent diagnostic and computerized adaptive remedial instruction system based on “Factor and Multiple” relevant indicators in elementary school for example by using knowledge structure analysis, and combining with the induction tool of Bayesian network. The system can diagnosis the wrong types of students instantly and can provide the real-time, on-line, right remedy teaching with computer animation so that the functions of evaluation, diagnosis, and remedy teaching can be achieved simultaneously. The results in the research are as follows: 1. The system based on knowledge structure analysis and combined with the evaluation mode built by Bayesian networks can be applied effectively to the diagnosis of students’ mistakes and sub-skills. 2. The computerized remedy teaching animation for students to achieve the significant progress level by using “the computerized adaptive remedial instructions” in this research is proved to be plausible. 3. Since the system can diagnosis the wrong types for each individual and give adaptive remedy teaching which are unavailable by traditional paper test, therefore the effect of teaching-by-individual and testing-by-individual can be certainly achieved.
