Abstract: Typical association rules consider only items enumerated in transactions, referred to as positive association rules.Negative association rules also consider the same items, but in addition, also consider negated items, i.e., those absent in transactions.Negative association rules are useful in market-basket analysis to identify products that conflict with each other or products that complement each other.They are also convenient for associative classifiers, classifiers that build their classification model based on association rules.Indeed, data mining using such rules necessitates examination of an exponentially large search space.Despite their usefulness, very few algorithms for mine this information have been proposed to date.In this paper, a fast and efficient algorithm MNAR is presented to discover negative association rules.Meanwhile, a method for calculating the support of itemsets is proposed.Experiments show that the MNAR algorithm is effective and feasible.

Key words: negative association rules, frequent itemsets, data mining