Design and implementation of students' score correlation analysis system

To make full use of students' score and discover relationships among courses, we designed and implemented a system based on web for students' score correlation analysis. The system can find the relationships among courses with students' score or grade rank of students' score. We use the Manhattan distance and correlation coefficient to measure the correlation between two courses. The system adopts 3-tier Browser/Server architecture, which composes of a presentation layer, a domain logic layer and a data access layer. The system can draw scatter plot, calculate Manhattan distance, calculate correlation coefficient and mine association rules with students' score or grade rank of students' score. There are two kinds of correlation coefficient in the system: the Pearson and the Spearman. In order to obtain association rules meeting user requirements, the minimal support and the minimal confidence in association rules mining can be set conveniently. The analysis results are displayed in form of tables and graphics. Graphics are drawn in the Canvas element of HTML5 with JavaScript. With the help of the system, we can find relationships and association rules among different courses.