The Personalized Course Recommended Applied Research Based On The Association Analysis Of Students' Scores

In recent years, the research of association rules mining has gradually developed in teaching, It is a hot research field in educational information about how to dig out the effective association rules from student data by use the association analysis and apply it into education and teaching. At the same time, due to the shortcomings of the blindness of course selection for students in the college course selection system,individualized course selection in colleges and universities has also presents a developing trend gradually.This paper studies the personalized course recommended based on the association analysis of students' scores, the main research work and contribution are as follows:Firstly, this paper proposes an improved NewApriori algorithm based on the classical Apriori algorithm of association rule mining, and compares the time performance of the NewApriori algorithm by experiments, and carries out the association rule recommendation experiment based on the NewApriori algorithm. The algorithm adopts the strategy of candidate set pruning. During the process of scanning the transaction database, the algorithm has improved the traditional method of scanning all the databases, pruned the set of candidate sets with the least degree of support, reducing the size of the data sets that need to be scanned. Therefore the efficiency of the algorithm for association rules mining has improved. The experiment use the 13140 student scores data of Central China Normal University as the transaction database, and the Apriori algorithm, the AprioriTid algorithm, the FP-Growth algorithm and the NewApriori algorithm proposed in this paper are separately run and compared. The experimental result shows that the NewApriori algorithm is better than the Apriori algorithm, the improved AprioriTid algorithm and the FP-Growth algorithm in time efficiency.Secondly, in this paper,the improved association rules mining algorithm based on NewApriori algorithm is applied to student computer course selection data and computer achievement data, and the association rule recommendation based on NewApriori algorithm is realized. The computer courses between which are promoted each other are analyzed from the experimental results, as well as the association relationship between the computer courses.Thirdly, this paper presents a personalized course recommendation algorithm model based on a hybrid method of the improved association rule mining and collaborative filtering. The recommendation module use the improved NewApriori algorithm to mine the association rules and implements the association rules recommendation, and the user-based collaborative filtering algorithm is the main part of the algorithm . In addition, the hybrid algorithm adds the weight to the recommendation result of the user-based collaborative filtering and association rules recommendation. It has solved the problem of data sparsity and cold start partially. The experiment use the student scores data of Central China Normal University as the test set, and then separately run the collaborative filtering algorithm, the association rule recommendation algorithm based on the improved NewApriori algorithm and the hybrid recommendation algorithm of the association rule mining and cooperative filtering based on the improved NewApriori algorithm to compare the quality of each algorithm. According to the results of the experimental results, the quality of the improved hybrid recommendation algorithm is better than that of the collaborative recommendation algorithm and the association rule recommendation algorithm based on the improved NewApriori algorithm.