Mining uncertain data for frequent itemsets that satisfy aggregate constraints

Abstract

Many existing algorithms mine transaction databases of precise data for frequent itemsets. However, there are situations in which the user is interested in only some tiny portions of all the frequent itemsets, and there are also situations in which data in the transaction databases are uncertain. This calls for both (i) constrained mining (for finding only those frequent itemsets that satisfy user constraints, which express the user interest) and (ii) mining uncertain data. In this paper, we propose a tree-based algorithm that effectively mines transaction databases of uncertain data for only those frequent itemsets satisfying the user-specified aggregate constraints. The algorithm avoids candidate generation and pushes the aggregate constraints inside the mining process, which reduces computation and avoids unnecessary constraint checking.