基于认证数据的学生上网时间特征分析

ANALYSIS OF STUDENTS’ ONLINE TIME CHARACTERISTICS BASED ON AUTHENTICATION DATA

为研究高校本科生上网时间特征,将学生上网认证数据转换成学生上网时长向量集.利用K-canopy算法去除离群点,并通过指标投票机制得到最佳聚类个数;利用K-means算法分别对工作日和周末上网时长向量集进行聚类,将工作日向量集聚为6个类、周末向量集聚为5个类;分析聚类结果,得到各类学生的上网时间特征、学生个人的上网时间特征和各专业中各年级学生上网时间特征.学生上网时间特征可为专业课程时间安排、学生管理等工作提供参考.

For studying undergraduates online time characteristics,the online authentication data of undergraduates were transformed into the vector set of their online time.K-canopy algorithm was used to remove outliers,and the quota voting mechanism was adopted for selecting the optimal number of clusters.The K-means algorithm was adopted to cluster the vectors of the time spent on the Internet on weekdays and weekends respectively.The vectors of the weekdays were clustered into 6 classes,and the vectors of weekends were clustered into 5 classes.By analyzing the clustering results,we can acquire online time characteristics of all kinds of students,a single student and students in different grades of a certain specialty.All those online time characteristics can provide reference for time arrangement of professional courses and student management.