Based Digital Twins Students Learning Behavior Management

Although data mining has been successful in the business world for a period of time,but it is in the higher education in the data use is still relatively new,namely its use to recognition and extraction from the data of new and potentially valuable knowledge.Use the purpose is to develop a model of data mining,the model can come to the conclusion that related to students’ academic success and master the activities of the law students with better management students.In the process of predicting the behavior of students,compares the different methods of data mining and technology,and application of the students from the university’s collection of personal,learning behavior,and campus trajectory data.Discussed in this paper,and use the basic machine learning and student performance prediction in this paper,the depth of the learning algorithm and campus track prediction,students will also discuss the continuity between data and machine learning.Hope the study will help the students themselves,teachers,education industry people and even parents to know more about forecasting the student’s final grade,and the campus life rule,at the same time can timely found abnormal behavior of students and warning,can also help them to take preventive measures.The main research content is as follows:In student performance prediction,a new CABLSTM deep learning architecture is established,which inputs the original signal to CNN to obtain feature information,and then uses the ABLSTM model to convolute the extracted features,and normalizes the time-frequency data in the regression layer to make performance prediction.In Student Campus Trajectory Prediction,a new method for next location prediction is introduced by combining students’ individual campus activity trajectories with the location preferences of students with similar daily activity patterns.Will find,the method through experiment contrast than traditional machine learning methods and improve the method of before,in this paper,the model has relatively lower error and higher accuracy,it is more convenient for the school management students,improve teaching environment provides support,the management of education and school better data mining school has important application significance.