Enhancing Association Rule Mining with the CGRG Algorithm: A Cluster-Based Approach

In the digital era, businesses increasingly rely on data analytics to derive actionable insights and boost revenue. Association rule mining is a vital technique for uncovering relationships between products, providing valuable insights into customer spending habits and product perception. However, existing algorithms often generate an overwhelming number of association rules, making it difficult for users to identify the most meaningful ones. This paper introduces the CGRG (Clustered Graph-Based Rule Generation) algorithm, a novel approach designed to produce a concise set of high-value association rules by leveraging a clustered minimum spanning tree framework. The CGRG algorithm generates two distinct classes of rules: bi-cluster rules and intra-cluster rules. By utilizing objective metrics such as support, confidence, and lift, the paper evaluate the quality of the rules generated by CGRG. Comparative analysis with the widely-used Apriori Algorithm demonstrates that CGRG consistently identifies the high-value rules within the extensive ruleset generated by Apriori, significantly reducing noise and improving usability. the results highlight the effectiveness of the CGRG algorithm in generating a compact, actionable, and high-quality ruleset, making it an invaluable tool for businesses aiming to optimize their decision-making processes and enhance customer insights.