An Optimization Study of Information Visualization Techniques in Educational Management Systems for Student Data Analysis

Abstract

Constructing a link between students’ behavioral data and real life can structure innovative college education and management ideas to more effectively guide and predict students’ behavioral development. In this paper, we design a student behavior analysis system to integrate relatively complete student data using preprocessing. Then the K-Means clustering algorithm based on information entropy and density optimization carries out clustering analysis of various types of student behavior data, and then uses the improved Apriori algorithm to mine the association rules between student behavior and performance. Applying the system of this paper in a university, it is found that the number of consumption and the amount of consumption are more concentrated and the range of fluctuation is small for students with good and medium grades. When the student’s life habit is often early, but irregular diet nor regular exercise, then the student has 84.068% probability of average overall quality. This paper provides a framework for dividing and representing various types of behavioral data of students using category symbols.