Development of a Causality Least Association Rules Algorithm Tool Using Rational Unified Process Methodology

Abstract

Among the most crucial research areas in data mining is association rule mining (ARM). Rules are classified into two types: frequent rules and least frequent rules. Extracting the least association rules is more difficult and always leads to the “rare item problem” quandary. The rules with the fewest items are known as the “least association rules.” However, most data mining tools favour frequent association rules over the least frequent association rules. Furthermore, the process of extracting the least association rules is more difficult. Therefore, this paper proposes and develops Causality Least Association Rules Algorithm Tool (CLART) using the Rational Unified Process (RUP) methodology and the C# programming language. The results showed that CLART is workable, and the proposed algorithm also outperformed the existing benchmark algorithm. In addition, CLART is a dedicated tool that is freely available and can be used to extract the causality least association rules from the benchmarked datasets.