A Review of Constrained Association Rule Mining

Pooja Dubey; R. K. Gupta

doi:10.37591/jowet.v2i2.448

Abstract

Market basket analysis is a topic of concern when Apriori was developed. By the time, the algorithms are evolving and focusing on reducing complexity, number of database scans, using certain checking to generate only useful rules. For generating rules, firstly by support value, algorithms can extract frequent itemsets. After specifying confidence certain rules are generated. Because of generating buying pattern, a large number of areas are using association rule as cross marketing, profit loss prediction and store management. These benefits of association rule mining are still need to be updated. More user specified checks can be incorporated by constraints. Constrained association rule mining is not only about applying constraints on the given dataset, but also a systematic way of constraints imposition and mining results based on novel patterns. Specifying taxonomies, non-uniform support constraints, and user specified constraints, small number of rules will be generated, so extratcted rules can be more useful and can be of importance.Keywords: Constrained association rules, generalized association rules, taxonomy based constraints, non uniform constraints Cite this ArticlePooja Dubey, R. K. Gupta. A Review of Constrained Association Rule Mining. Journal of Web Engineering & Technology. 2015; 2(2): 18–22p.