Using Multiple Minimum Support to Auto-adjust the Threshold of Support in Apriori Algorithm

Nowadays, Data mining becomes an important research domain, aiming to extract the interesting knowledge and pattern from the large databases. One of the most well-studied data mining tasks is association rules mining. It discovers and finds interesting relationships or correlations among items in large databases. A great number of algorithms have been proposed to generate the association rules, one of the main problems related to the discovery of these associations (that a decision maker faces) is the choice of the threshold of the minimum support because it influences directly the number and the quality of the discovered patterns. To bypass this challenge, we propose in this paper an approach to determine how to auto-adjust the minimum support threshold according to data by using a multiple minimum support. The experiments performed on benchmark datasets show a significant performance of the proposed approach.