A Study On The Correlation Between Student Behavior And Academic Performance Based On Data Mining

Nowadays,the fast development of the Internet is promoting our society to step into the era of big data from the information age.With the annual International Conference on Education Data Mining organized by the International Working Group had successfully held in 2008,both the domestic and foreign research on education data mining have experienced a blowout and have became a major direction in the application of big data of the field of education.What’s more,as for education evaluation,student’s study achievement acts as a significant indicator in evaluating the education quality of school and serves as an examination of student’s achievements and teacher’s teaching effects.At present,there are many researches on data mining between students’ behavior habit and their study achievements.However,most of these researches about behavior habit focus on discussing the student’s learning behavior habit in class and on the Internet,while few of them focus on developing the student’s learning behavior habit in daily life and after class.Therefore,this dissertation collected 397 valid questionnaires from a high school in Yangzhou city by means of questionnaire survey.Firstly,this research exerts the decision tree classification algorithm of data mining to model and analyse the relevance between student’s life behavior habit(night sleep,afternoon nap,diet and exercise)and student’s learning behavior habit after class(preview and review).It gets the results that the relevance degree between leaning behavior habit and study achievement is stronger than the relevance degree between life behavior habit and study achievement.In addition,the relevance degree between these six behavior habits and study achievement can be ranked in a strong to weak order: review,preview,night sleep,diet,afternoon nap and exercise.Secondly,this research uses the K-Means clustering algorithm in data mining to group students,which provides scientific basis for teachers to manage students.Through the comparison between two data mining models,a relatively consistent result of two models comes out.This indicates that it is feasible to analyze the relevance between student’s behavior habit and their study achievement through data mining.Last but not least,based on the results of two mining,this dissertation offers suggestions for students to improve their study effecct and for teachers to enhance their daily teaching and the management of students form the perspectives of students and teachers.