The Research And Implementation Of Grade Data Mining And Analysis System Based On Multi-strategy

With the enrollment expansion of most universities, the number of the students increases sharply, which makes the problems of present educational administration systems more serious. Especially, the module of students'grade analysis only does such simple things as statistics analysis or data sorting. As a result, a great deal of useful information hidden in mass data can't be utilized efficiently. Therefore, it deserves for further research of university that how to represent these information as knowledge in order to support school decision-makers'decision and instruct teaching and improve educational administration.Data mining technology 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 comes up with multi-strategy analysis method and combine data mining technology with statistics analysis to extract useful information from mass grade data. Multi-strategy means as follows: Utilizing classifying data mining methods based on decision tree to analyze the data in grade database. A grade decision tree is generated to show directly a position of grade according to different computing methods and to support estimate. At the same time, utilizing classification method based on summing-up principles to do such things as grade query analysis and prediction and contrast analysis to realise automatic generation of grade analysis report, test paper's quality assessment report and quality analysis table which plays an active role in improving teaching and test paper's quality. A system is developed which runs experimentally to anticipated effect.At last, a summarization is made and other data mining methods'future application into grade analysis is prospected.