The Research And Application Of Clustering Analysis And Association Rules In The Achievement Analysis

With the increasing growth of university enrollment expansion plan, the number of students has increased greatly, brings a great challenge to school teaching. According to the development of information technology, there are a lot of information systems and databases in the school and they are used to store student information. A large amount of data is accumulated in these systems and database, but the information behind the large data do not be used effectively. As a knowledge discovery technology that data mining can extracts valuable rules or schema, it can help us make effective use of student informationSchool holds terminal examination every year, so the college accumulates a lot of test data.We just store and query the mass data. However, the information that maybe valuable behind the score data has not been analyzed and used effectively. So the study makes an application of data mining technology in the student achievement analysis of "Fundamentals of Computer" and dig out the useful knowledge for teaching from the score data.Based on the learning and understanding the theory of data mining, this study focuses on the clustering analysis and association rule mining technology.Then, it makes a detailed study of the classic K-Means algorithm and Apriori algorithm, and designs the appropriate K-Means algorithm and Apriori algorithm for "Fundamentals of Computer" data, applies them on the achievement analysis of "Fundamentals of Computer". First of all, we select the data source, a university2011-2013grade student achievement of the course" Fundamentals of Computer" is the research object, and then perform data preprocessing, like data integration and data reduction. Second, we use K-Means algorithm, and choose the k-value and initial clustering centers according to the distribution characteristics of student achievement. Then we analysis the clustering results and obtain the information which is the useful and meaningful. Third, After the clustering analysis, to further determine the credibility of the clustering results and the relationships among student’s departments score of each question type and student achievement, the thesis use Apriori algorithm for the data and find frequent sets, then find the association rules and obtain the factors affecting the student achievement.Through the idea that combines the clustering analysis and association rules, we mine the achievement of "Fundamentals of Computer" and obtain useful knowledge for teaching and offer the decision-making basis for teachers’ teaching, expecting to use this method in other courses.