earticle

영어

For the past decades and until now, association rule mining is one of the most prominent research topics in data mining. However, the main challenge among public or private practitioners is to find the interesting rule from data repository. As a result, many efforts have been put forward to explore this rule by applying several methods and interesting measures. Therefore, in this paper, we introduced an enhanced association rule mining method namely Significant Least Pattern Growth (SLP-Growth), where the algorithm embeds with two interesting measures called Critical Relative Support (CRS) and Correlation (Corr). The experiment uses the dataset that contains the records of preferred programs being selected by post-matriculation or post-STPM students of Malaysia via Electronic Management of Admission System (e-MAS) for the year 2008/2009. The experimental results show that the SLP-Algorithm with the embedded measures can successfully in categorizing the association rules. In addition, this information can be used by educators and higher university authority personnel in the university to understand the programs’ patterns being selected by the students. More importantly, it can assist them as a basis to offer more relevant programs to the potential students rather than by chance technique.