Analysis For Students' Performance Based On Data Mining Technology

The 21st century is the era of knowledge-driven economy, and it is an overall information era for each field of the society especially in education. The level of education information and effective occupying percentage of digitized teaching resources has become an important sign which weighs a country or the regional educational modernization process. The education information is a lifeline of education in the 21st century.Teachers can only obtain superficial information by simple statistics or sequencing functions, because of lacking the information consciousness and technology, the information hidden in the large amount of data has never got application. Many schools are considering the following question: how to utilize the data again, how to translate the existing management data into the knowledge suitable for using, how to improve management decision-making in school and teaching quality. Data Mining is a useful way to solve the lack of the information. Its essence is the process during which the unknown, hidden and useful information is picked up. It functions prominently in mass data analysis and has been applied successfully in many fields.After analyzing and comparing the data mining technology deeply, this dissertation combine data mining technology with statistic analysis:Utilizing statistic analysis based on summing-up principles to do such things as grade contrast and analysis to realize generation of test paper's quality assessment.At the same time, data mining technology has all kinds of mode, such as association analysis, classification and forecast. Each mode has its own emphasis, among them, there are some already studied modes have much more research outcome. But put some of them as data pre-processing for others to mine the data, at present, related applied research has not appeared yet. This dissertation utilize attribute reduction algorithm of the theory of rough sets and clustering analysis for data pre-processing to discretize the attribute value and reduce the redundant attribute on questionnaire. Then association rule is applied in order to find out the relationship between students'performance and their behavior during the academic life.Most of the previous using tools in related applied researches are selected with the statistic analysis functions of the SPSS or SAS software for statistic analysis and Visual Basic,Visual C++,Visual Foxpro,Delphi or the data mining functions of Microsoft SQL Server 2000 and so on for data mining. This dissertation implements analysis for students'performance based on MATLAB conformably.