A Framework for Synthesizing Arbitrary Boolean Queries Induced by Frequent Itemsets

Abstract

Frequent itemsets determine the major characteristics of a transactional database. It is important to mine arbitrary Boolean queries induced by frequent itemsets. In this paper, the author proposes a simple and elegant framework for synthesizing arbitrary Boolean queries using conditional patterns in a database. Both real and synthetic databases were used to evaluate the experimental results. The author presents an algorithm for mining a set of specific itemsets in a database and a model of synthesizing a query in a database. Finally, the author discusses an application of the proposed framework for reducing query processing time.