Efficient Associate Rules Mining Based on Topology for Items of Transactional Data

A challenge in association rules’ mining is effectively reducing the time and space complexity rules with predefined minimum support confidence thresholds from huge transaction databases. In this paper, we propose an efficient method based on topology of itemset for associate To do so, deduce a binary relation itemset, construct quotient lattice according to transactions itemsets. Furthermore, prove that all closed itemsets are included topology, generators or minimal every can be easily obtained element lattice. Formally, represents more general associative relationship among items databases, displays hierarchical structures itemsets, provide us approximate any template itemset. Accordingly, algorithms generate Min-Max reduce generalized lower approximation upper template, respectively. The experiment results demonstrate proposed alternative