Abstract | ||
---|---|---|
Nowadays, a large amount of valuable data have been accumulated. According to the big data from the management system of university, we attempt to subdivide students' behavior into different groups from various aspects, so as to identifying the different groups of students. Given this, this paper can get the characteristics of students from different groups. In this way, universities can know students well and manage them reasonably. First, in order to solve the segmentation of student behavior, this paper presents a set of description index system of student behavior and the segmentation model of student behavior based on clustering analysis. Meanwhile, in order to obtain more accurate clustering results, the traditional KMeans clustering algorithm is improved from the selection of the initial clustering center and the number of clusters. In addition, the improved method is parallelized on the Spark platform and applied to subdivide student behavior into different groups. Finally, experiments are conducted to verify the reliability of the results. |
Year | DOI | Venue |
---|---|---|
2017 | 10.1109/CIS.2017.00116 | 2017 13TH INTERNATIONAL CONFERENCE ON COMPUTATIONAL INTELLIGENCE AND SECURITY (CIS) |
Keywords | Field | DocType |
Campus big data, Student behavior, Clustering, K-Means, Spark | Data modeling,Algorithm design,Spark (mathematics),Segmentation,Computer science,Index system,Artificial intelligence,Cluster analysis,Management system,Big data,Machine learning | Conference |
Citations | PageRank | References |
0 | 0.34 | 0 |
Authors | ||
4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Dong Ding | 1 | 0 | 0.34 |
Junhuai Li | 2 | 39 | 16.44 |
Huaijun Wang | 3 | 20 | 13.02 |
Zhu Liang | 4 | 0 | 0.34 |