부산대학교 도서관

Evaluating students' achievements which including many subjects is a great challenge. It seems not so accurate that simply adding these scores and ranking, because it is difficult to distinguish which one is a student good at and which one is not. Data mining is a good way to solve this problem. In this paper, students' achievements are appraised based on factor analysis and cluster analysis. First, common factors are extracted from scores of multitudinous subjects. Then factor scores and comprehensive scores can be computed. After that, all students can be segregated into several clusters by cluster analysis based on factor scores. The result shows objective synthetical evaluation of students, which will benefit education in the future.