The Data Minting Appliction In Drop-Out Self-Study Students Analysis System

With the rapid development of the global informationization, mass data has been created and accumulated. Since useful information need be obtained from mass data for people to make well-informed decision, data mining technology has become an urgent demand. It is the same case in modern education examination field, where the data or information of many students has been saved by the computer. Numerous information rules about education examination are existing in these data and information, which also needs data mining technology for studying and applying these data. At the beginning, this dissertation defines the concept of Data Mining and Data Warehouse, which is the necessary environment of Data Mining. Then it discusses the relevant issues about data mining covering its arithmetic, technique, real-world process and its architecture. From those, the necessity of Data Mining in analyzing the drop-out student of Beijing Examination Institute's self-study examination can be proved. Also, we choose the decision tree's Supervised Learning In Quest (SLIQ) arithmetic as the arithmetic of actual work at the real demand and detailing the data preparation and the way of data splitting used in Data Mining arithmetic. With the combination of real drop-out student analysis system and data mining demand, we build up a Data Warehouse on examinees'drop-out system data mining. Based on SLIQ arithmetic,we use the examinee'data in data warehouse to implement the training of decision model through the analysis of examinee's public course examination mark, times of examination and graduation situation. A decision tree model was built about student's drop-out situation, through which to provide Beijing examination institute with the theoretic support for their policy about self-study student's drop out.