ITL-MINE: Mining Frequent Itemsets More Efficiently.
Raj P. Gopalan, Yudho Giri Sucahyo · 2002
The discovery of association rules is an important problem in data mining. It is a two-step process consisting of finding the frequent itemsets and generating association rules from them. Most of the research attention is focused on efficient methods of finding frequent itemsets because it is computationally the most expensive step. In this paper, we present a new data structure and a more efficient algorithm for mining frequent itemsets from typical data sets. The improvement is achieved by scanning the database just once and by reducing item traversals within transactions. We present performance comparisons of our algorithm against the fastest Apriori implementation and the recently developed H-Mine algorithm. These results show that our algorithm outperforms both Apriori and H-Mine on several widely used test data sets. 1.