Design And Implementation Of Academic Warning And Social Analysis System Based On Student Campus Data

The rapid development of data mining technology and artificial intelligence algorithm has laid a solid theoretical and applied foundation for us to mine valuable information from massive data and make use of it.With the vigorous development of wireless network and smartphone,and the popularization of campus digital management system,we can obtain and analyze the trajectory and behavior data generated by students' rich and colorful campus life,excavate valuable information and provide data-driven guidelines for more scientific and effective management of students.Based on campus multi-source heterogeneous data,combined with Wi-Fi positioning technology,trajectory data mining technology,social network analysis technology,machine learning and in-depth learning technology,this paper designs an algorithm to predict students' performance from three levels:personal portrait,social behavior portrait and space-time trajectory model,and designs and implements an academic early warning and social analysis system.Firstly,multi-source heterogeneous data including Wi-Fi probe location,campus network usage data and campus card usage data are preprocessed and fused based on temporal and spatial information to construct students' campus trajectory.Then,algorithms are designed to construct students' individual user portraits and collective user portraits,and the correlation between portraits and students' academic performance is demonstrated and analyzed.Then,based on machine learning and in-depth learning technology,the algorithm combines personal portrait,social portrait and trajectory model to predict students'academic performance,and uses the accuracy,recall,accuracy and AUC to measure the prediction results.Finally,we design and implement an academic early warning and social analysis system to provide the functions of student portraits,social relations and screening of abnormal students.