A distributed framework of student data early warning system

Abstract

With the deepening of research on big data, various intelligent technologies have been widely used in the digital construction of campus. But the current campus students early warning system can not achieve high accuracy results. It is necessary to collect big data of students and perform online analysis with advanced techniques to timely predict problems of learning performance and activity abnormality. In this paper, the campus big data early warning system is designed based on the distributed framework. It integrates the data collection, storage, mining and analysis, visualization and message notification. The system consists of six layers of data application structure and one security control platform, and uses multiple linear regression algorithm and logistic regression algorithm to predict the abnormal behavior of student, Hadoop and Spark are used to build a distributed framework to monitor student abnormalities.