Research On Curriculum Correlation And Visualization Based On Association Rules Mining

Until now,association rule mining algorithm is still one of the core points in the field of big data research,and it is applied in various fields.Among them,the combination of education and data mining is the most common way,which can better provide targeted learning and teaching plan for students,teachers and students.In the massive information of students’ course selection,hidden information with certain value is often hidden,which needs to be further discovered artificially.At present,there are many researches related to association rules mining algorithm,mainly focusing on how to find frequent item sets and how to prune.There are many algorithms in this aspect,but most of these algorithms deal with and analyze the whole data set.When the number of data sets is large,mining the whole data set will reduce the efficiency of the algorithm.Although the results can be obtained by related algorithms,the efficiency is not high and the results are not targeted.In order to solve this problem,this paper proposes an algorithm.Firstly,K-means algorithm is used to cluster data sets to generate a specific number of clusters.The improved Top-K algorithm combined with correlation coefficient is used to find the association rules in each cluster.The combination of clustering and improved Top-K algorithm can classify the data set first,and then conduct mining analysis.Through this idea,the accuracy and efficiency of the whole algorithm can be improved.The prediction model of K-means algorithm is constructed by BP neural network.The final experiment shows that the integrated algorithm can be used to mine and analyze course association rules.According to the user’s course selection data set,the algorithm mines the association rules in the courses selected by the same type of users.Improve the pertinence and accuracy of the original algorithm,and compared with the original algorithm,the running time is shortened by 14%.Finally,in order to intuitively analyze the mining results,based on the experimental data,this paper analyzes the current research status of data visualization,and according to the characteristics of association rule mining results,puts forward a display method which can be used to display the results of course data association rule mining,and uses this display form to display the experimental results of data visualization.