The Research On The Application Of Association Mining In Scientific Research Management System

With the rapid development of database technology and the extensive applications of database management system, more and more data is accumulated. Important message can be found by data mining technology, which can provid scientific basis for higher level analysis. After years of development, data mining algorithm has become increasingly mature, stable, and easy to understand and operate, then the data mining research has gradually shifted its focus on how to applicate. Currently, data mining technology has accumulated a lot of successful experiences in the commercial, and financial fields. But applications for education are still very few. In recent years, with the constant expansion of universities, the number of students substantially increases, which has brought severe test to the teaching, scientific research management of universities. We found that making use of massive teaching, scientific research information which has been accumulated for years, to discover scientific basis for enlightenment through data mining, which can not only guide universities seek reasonable mode of instruction, but also guidance teachers and students for effective research work, thus colleges and universities can make a good direction healthily and steadily.This paper introduces the Data Mining into the management of scientific researching. Based on a university's scientific research management system, this paper uses association mining technical to find the key elements in scientific researching, and the association about research elements and research fruit. The mining of these information will promote the actualizing of scientific researching policies.This paper summarizes the background and advantage of data mining, as well as the significance of data mining in the scientific research management of colleges and universities firstly, and then discusses the theory of data mining, association rules and the ideas of main algorithm, analyzes the classic Apriori algorithm and its existing problems as well as the basic solutions. After that, this paper proposes Multi-Dimensional Apriori algorithm which is designed specially for the mining of this paper; Then describes the structure of scientific research data mining system, defines subject-oriented mining tasks, including: mining the data about research projects, mining the data about papers, mining the data about academic writings. The association mining process is implemented by programing, a number of stimulating association rules are found, interpreted and analyzed. Finally this paper also presents some follow-up research work in future.