Analysis Of Behavioral Characteristics Based On Student's Personal Big Data

With the educational informationiziton construction being strengthened and completed,the data of students' daily life and learning behaviors are recorded,and stored in the Digital Campus Platform,which has initially formed the students'personal big data environment,characterized with large-scale and multi-type.It has attracted extensive attention from the researchers and the university administrators.Most scholars' studies focus on the correlation analysis between students' behaviors and academic performance,which based on a single or few types of behaviors' data.However,the problem,which exists in the most studies,is the number of indicators of students' behaviors is small,and the categories are not comprehensive,in addition,there is a lack of overall analysis about students' growing and changing.In response to the above questions,this study try to analysis and explore the growing and changing of college students by using student' personal big data,and to excavate the characteristics and correlation behind students' different behaviors,and to assist to enhance the school's detecting ability about students' growing and changing.The research contents of this paper mainly include the following three aspects:1.The students,data is classified and summarized from three fields,which include students5 basic information,campus learming and campus life,and then building the students' personal big data behavior analysis model by extracting behavioral feature from four aspects,which mainly includes student Campus Consumption,Online Learning,WiFi Network Using,Courses and Achievements.In addition,analysizing and mining of students' consumption behaviors?data deeply to explore students' diets and consumption levels.2.A data analysis platform for students' personal big data is designed and built,which is based on Hadoop,a distributed computing framework.3.Based on the behaviors data generated by undergraduates of four different grades in a university in Wuhan,which including 6 aspects about students' basic information,grades,consumption,library access,online learning,and national scholarship winners,from school year 2014 to 2016.Students from different grades are analyzed in an overall way to explore the differences and commonalities of the behavioral performance.And mainly focusing on the analysis of excellent students'behavior characteristics in college,as well as the correlation between academic performance and behaviors,and drawing "student portrait,for individual students.Through data analysizing,the following results were found:1)From the consumption data of all students,the number of meals consumed by students during school is decreasing year by year,and the number of breakfast meals is decreasing year by year;the peak time of breakfast for freshmen is 7 to 8,which is one hour earlier than the all students' numbers of meals.2)From the data of all students'libraries,juniors have the most frequent access to the library,and the performance of sophomores entering the library is better.3)From the correlation analysis of student behavior and performance,there is a strong correlation between the student's academic performance and meal frequency,breakfast meal rate,and eating consumption level.The correlation is weaker in the student's academic performance and the frequency of learning in library,the choice of dining window.4)the lower and the more stable their consumption level,and the more diligent they are in their study,the better their academic performance and the nore likely they are to win the national scholarship.