Advancements in single-label associative classification: A systematic and comprehensive review

Association Rule Mining and Classification are important tasks in data mining. Using association rules has been proven to be a good approach for classification. Associative Classification is the integration of Association Rule Mining and Classification, very useful in several domains due to its high interpretability based on the Class Association Rules that the model offers to end users. While several surveys have reviewed this approach broadly, none have systematically focused on the single-label setting, which represents the methodological core of the field. This paper presents a systematic literature review of single-label associative classification from 1998 to 2025, conducted following EBSE guidelines and reported using PRISMA principles. We analyze major algorithmic families, compare their classification strategies, examine reported empirical results, and identify methodological gaps, like in standardization and interpretability. The review provides a consolidated perspective on single-label associative classification and outlines key directions for advancing interpretable rule-based classification, like the use of negative Class Association Rules in tie-breaking.