An Efficient Algorithm for Mining Top-Rank-k Frequent Weighted Utility Itemsets in Hierarchical Quantitative Databases

The problem of mining frequent patterns represents a fundamental challenge in data mining, particularly in the extraction of association rules. Recently, this problem has also served as input data for various intelligent systems, contributing to their development across different databases. Among these, mining top-rank-k frequent patterns on hierarchical quantitative databases is a significant problem that requires further research. In this paper, we propose the HNK_FWUI algorithm using the HNK-list structure to mine top-rank-k frequent patterns on hierarchical quantitative databases. By employing minimal database scans and early pruning strategies, the algorithm demonstrates greater efficiency compared to traditional methods that use data structures such as EDBV and EIWS in experimental evaluations.