Research And Development Of University Student Behavior Data Analysis System Based On Big Data Mining

In the modernization of higher education management,the level of information technology increased year by year,with the extensive use of campus card and the accumulation of major business system data in calendar years,forming a large campus data environment.It is mainly manifested in the characteristics of large-scale,multitype,high-speed and low-density value of students 'data.How to dig the student's card data effectively becomes the important content of improving the students' information management level.This study mainly studies the data of undergraduate students' onecard data and the behavior of students in the corresponding business system(learning behavior,life behavior,mental behavior).The data includes student's consumption data,school's hospital visited data,access data,library borrowing data,test scores,Internet time and other massive data.The data of the students 'learning,life and psychology are explored.The abnormal data of the students are collected and returned.And the digital campus and the intelligent campus are made full use of the students' school behavior data resulted in improving school's information technology level.Based on the data analysis system of students 'behavior,this thesis,based on the research work of the ideological and political education of capital university students under the Beijing Municipal Committee of CPC,"Data analysis and application of university students' card data under the perspective of large data",Mainly to complete the following:(1)Integrate the historical data of each major business system of the school,analyze the data of the students in school card,process the abnormal data.(2)Research the HDFS file system and MapReduce model of Hadoop,build the overall technology architecture of university student behavior data analysis system based on Hadoop technology,and use the computing model of MapReduce to mine the student behavior data.(3)The students' behavior data points are reduced,the relationship between different behaviors is combed and the students' "students' portraits" is painted.Clearly describing the situation of students in school,Correlation Analysis of student learning,life status and the psychological dynamics.(4)To construct the recognition model of college students with financial difficulties,and use the concept of membership degree of fuzzy evaluation method to combine the data of school card data and family situation questionnaire in large data analysis system to determine the students' membership level so that determining their poverty level.(5)Building student behavior's big data analysis system,analyze and summarize the behavior and characteristics of students,put forward constructive reference for the analysis of relevant departments,in order to analyze the behavior characteristics of students,and timely guidance of student behavior's comprehensive and healthy development.(6)To complete the big data analysis system in the use of the work of families with financial difficulties students.The use of the model to determine the way instead of the previous qualitative analysis and turn qualitative to quantification improved the efficiency of student work and the identify results' scientific and reliability.