ArmSquare: An Association Rule Miner Based on Multidimensional Numbered Information Spaces

Abstract

In this article, we propose a simple approach for association rule mining, which uses the possibilities of the multidimensional numbered information spaces as a storage structures. The main focus in the realization of ArmSquare is using the advantages of such spaces, i.e., the possibility to build growing space hierarchies of information elements, the great power for building interconnections between information elements stored in the information base, and the possibility to change searching with direct addressing in well structured tasks. The tested types of implementations of realized tool show the vividness of proposed approach.