Applying Data Mining Techniques To Vocational Colleges Enrollment

In recent years, with the development and widely using of Database systems,commercial and industrial areas have accumulated vast amount of data. However,extracting useful information has proven to be extremely challenging. And traditionaldata analysis techniques are not suitable because the massive size of a data set. Thus,data mining techniques are studied are developed in recent years, and it has beensuccessfully adopted in many areas.The higher education system of China has been changing rapidly those years. Itis moving from elicit education to popular education. And the number of vocationalcollege students grows especially fast, therefore vocational colleges have accumulatedvast amount of data. And current college entrance examination system andmanagement system has some drawbacks to some extent, education policy makershave been working to reform the college entrance examination. The quality ofeducation depends on the level of most of the students’ professional mastery.Therefore, for colleges and universities, in order to improve their quality of education,we must firstly have a thorough understanding of the students, and combine it with thecharacteristics of the college itself. Then use the combination of the two to developappropriate recruitment strategy, eventually form an education system with its owncharacteristics of a college.In this paper, we bring data mining technology in college enrollmentmanagement applications. By leveraging existing data mining techniques, such asassociation rules and decision tree algorithm, and combined with the characteristics ofcollege enrollment data, we achieve a viable data mining program, which is able towork for the college’s enrollment management to bring new improvement, so as toprovide useful guidance for the cultivation of students. Furthermore, in theimplementation, a hash method is proposed for the Apriori algorithm, a votingstrategy is adopted into the C4.5classification approach, to enhance theirperformance, therefore we believe these contributions will be of significance andworth existing.