An Application of Data Mining in Knowledge Points Correlation Analysis

Abstract

Obtaining the association among knowledge points of courses helps teaching activities. Mining the association rules needs to deal with two key issues. â‘  Data processing. â‘¡ Rule filtering. We proposed a method for knowledge point association analysis. In the data processing, we used the density peak clustering method on student data to eliminate the outlier data. Then we discretized the lost score of knowledge points by normal distribution. In rule filtering, we used the Apriori algorithm to mine the association rules which have high occurrence frequency, filtering out the meaningless associations caused by the difficulty factor of test problems exceeding the occurrence threshold of events. The knowledge point associations achieved above can guide teachers to arrange the teaching events with focuses, helping students comprehend the knowledge structure of whole course.