Research On The Application Of Decision Tree Algorithm In The Analysis Of Students’ Course Performance

With the advent of the information age,the development of educational informationization in China is constantly promoted.In the process of teaching implementation,there are all kinds of educational platforms rise rapidly,which makes the number of educational data increased continuously.Due to the limitation of data mining technology,relevant educational data can only be stored,queried,and mathematical statistics,its potential information value has not been paid enough attention;however,the promotion of educational information at this stage makes some of universities pay more attention to care about educational data.The potential value of education data is constantly explored.It can obtain useful knowledge from the data and information,find the relationship and rules between the data,which provides the appropriate decision-making basis for teaching link and teaching quality.In order to explore the factors influencing the performance of the online course "Application of Modern Educational Technology",a public compulsory course for full-time undergraduates in our University.This research summarizes previous studies in this field after studying a large amount of literature,and builds on this foundation I put forward my own new ideas for research in this field.At the same time,I applied data mining technology to the education and teaching work,and fully explored the relationship between the influencing factors of course performance.In order to apply relevant research to actual teaching,this research realized the early warning of course performance.The design and development of the system provide a scientific basis for the optimization of the course’s teaching strategy,the improvement of course grades,and the early warning of course grades.It also provides reference for the realization of other course grades mining.Specifically,this study mainly did the following work:(1)Based on the online learning platform of "Superstar Learning Link",this paper analyzes the data of online learning behavior of full-time undergraduate students in our University,and observes the information value of the data.(2)Using decision tree technology in data mining technology to implement in-depth mining of course learning behavior data,explore the relationship between various data,establish a CART decision tree model of the influencing factors of course performance,and analyze the factors that mainly affect the course performance from the model(Course video completion percentage,chapter study completion percentage,total number of participating discussions,exam completion percentage,teacher title),and it based on the model to summarize the decision tree generation rules to provide a scientific basis for education and teaching.(3)Rely on the decision tree generation rules to realize the design and development of the student’s course performance prediction system.Through the deployment of the developed system,the debugging and operation of various functions,the actual course learning behavior data is brought into the system for further testing to verify the influencing factors of the course performance.The rationality of the construction of the decision tree model and the effectiveness of the system development have realized the prediction of the undergraduates’ "Modern Educational Technology Application" course grades in our University,which provided a certain reference for the education and teaching of this course.