Formal Concept Analysis Based Association Rules Extraction

TL;DR

This paper introduces a novel method based on Formal Concept Analysis to extract more meaningful and relevant association rules, reducing the number of rules and enhancing their utility in decision making.

Contribution

It extends Formal Concept Analysis by generalizing itemsets and introducing a new relevance-based quality criterion for rule extraction.

Findings

Effective reduction in the number of rules generated.

Enhanced relevance and semantic value of extracted rules.

Improved decision-making support through more meaningful rules.

Abstract

Generating a huge number of association rules reduces their utility in the decision making process, done by domain experts. In this context, based on the theory of Formal Concept Analysis, we propose to extend the notion of Formal Concept through the generalization of the notion of itemset in order to consider the itemset as an intent, its support as the cardinality of the extent and its relevance which is related to the confidence of rule. Accordingly, we propose a new approach to extract interesting itemsets through the concept coverage. This approach uses a new quality-criteria of a rule: the relevance bringing a semantic added value to formal concept analysis approach to discover association rules.