Research And Implementation Of Data Mining Based Education Analysis System

In today's era of education,the vast education data brought by the development and popularization of college education informatization is most of the idle and abandoned,the status of developing educational decisions entirely relying on empirical is very incompatible with the new educational environment.A lot of education data needs to be highly utilized and analyzed.The subject accord to the actual education environment in colleges and universities,in view of the undergraduate education system and the training plan,put forward to three data-analysis themes:1.Use classification algorithms to model student achievement from the professional courses grade,predict whether one's profession ability meets the standards of training programs through his pre-basic courses.2.According to the information collected in the process of teaching and student's performance of various stages,explore factors that influence the teaching effect and extract the students' learning status,in order to specify the personalized coaching and reinforcement learning programs.3.Analysis the association from professional course select records of a large number of students,explore the habits and rules of course selection behavior,to allocate education resources reasonably and improve overall teaching level.This paper use knowledge of education data mining to convert these needs into such problems as classification,clustering and cluster problems,use Weka as the experimental platform,uses C4.5,SVM,Logistic regression,RIPPER,DBSCAN,FP-Growth algorithms algorithm respectively to model and analyze each problem,use cross-validation,plotted ROC curves and two-dimensional scatter plots,etc.on the cluster model to evaluate the results,and then analysis,create and execute the research plan and verify the validity of the scheme.Based on the findings and conclusion of the experiments,the paper designs and implements the education data analysis system.Create a new,dedicated data source module based on MySQL database;makes use of Weka open source API to build the algorithm modeling,evaluation and result output modules;and complete related tests.