Application Of Decision Tree Algorithm In Student Performance Prediction

Since the 21 st century,human society has entered the information era in an all-round way.Internet information technology has penetrated into all aspects of the society and has profoundly affected the way of people to live,study and work.Social progress and economic development are increasingly dependent on Internet information technology and information resources.In the field of education,the rapid development of information technology has become an important means to improve the quality of teaching.It has also contributed to a better reform of teaching models,ways and methods,enriched teaching resources and improved the efficiency and quality of student learning.The amount of education-related data are increasing rapidly,which helps the study of education,such as teachers understand the effectiveness of student learning and promoting teaching reform,and provides abundant data support.There are many original educational administration management systems and various online teaching platforms under construction.A large number of data about students learning are produced in the database of each system platform.For example,student personal information(including student ID,name,gender,major,class,etc.),course examination grades,course test results,on-line examination results,number of online learning,time points and length of learning,homework completion and so on.At present,these data only exist in various network learning platforms and educational administration departments.The data has not really played its due role.This paper mainly uses the CART algorithm and the random forest algorithm.Through analyzing the students' basic information,the course examination results and the students' learning data and the related characteristic attributes of the on-line learning,the decision tree algorithm is used to generate a decision tree,which produces the classification rules,and find out the main factors influencing student achievement from the Classification rules.It provides the powerful data support for teaching management departments and teachers to improve teaching quality.Through the steps of data extraction,data preprocessing,decision tree construction and decision tree optimization,this paper constructs a prediction model of student course scores,and puts forward the method of model realization;At the same time,the importance of the relevant attributes of the model is analyzed,and put forward some corresponding strategies to improve the level of online learning platform and teaching methods.The classification performance of the two algorithms is analyzed and compared,and how to improve the robustness of the algorithm.Finally,a student achievement prediction system is designed by using java to realize the classification and prediction of student course examination results.