Research And Application Of Grade Prediction Model Based On Decision Tree Algorithm

Grade prediction and early warning is to predict students' final grade through various data reflecting students' learning situation generated in the course of learning,and to give early warning of the possible phenomenon of low academic performance.Therefore,it is of great practical significance to study the grade prediction model for improving the quality of teaching and reducing the rate of students' failure.Aiming at the course C Language Programming of Xi'an University of Technology,this paper studies a grade prediction model based on decision tree algorithm,and constructs a course grade early warning system with this model.Firstly,the learning data of the class of 2017 and the class of 2018 in the School of Computer Science of our university about the course C Language Programming is collected,including the lab score,the number of solving questions,the chapter test score and the final grade.The collected data is preprocessed,the learning behavior is analyzed and the feature is extracted.The performance of the final stage grade prediction model constructed by ID3,C4.5 and CART decision tree algorithm is evaluated through experiments.C4.5 is selected for the algorithm of the final stage grade prediction model constructed in this paper.Secondly,by analyzing the influence degree of each feature on the final grade,the weight is introduced to improve the C4.5 algorithm,and the performance of the final stage grade prediction model before and after the improvement of the C4.5 algorithm is evaluated through experiments.The experimental results show that the accuracy of the improved C4.5 algorithm in predicting the passing or failing of the final grade can reach 92.49%.More recently,in order to achieve real-time grade prediction and early warning,the stage grade prediction model is constructed based on the original C4.5 algorithm,which can predict the final grade at each stage of the course learning.Finally,according to the functional and performance requirements of the course grade early warning system,a course grade early warning system of C Language Programming is designed and implemented by using the grade prediction model.The course grade early warning system implemented in this paper can provide teachers with the functions of real-time understanding of the learning situation of class students,the prediction results of course grade and early warning notification to students in crisis.It can also provide students with the functions of real-time understanding of individual learning situation,the prediction results of course grade and viewing the early warning notification.