An Efficeint Algorithm to Mine Negative Regular Itemset using Vertical Database

Abstract

Recently, mining negative patterns has received some attention from the researchers because there are huge numbers of such patterns that can be derived from transactional databases. Mining negative patterns are relatively tough compared to positive patterns. They are logically less in number and the constraints to find them are often complex. In this paper we propose an algorithm (RN2I) to find all possible itemsets of size two which are regular and negative from a transactional database using vertical data format. Our experiment results show the efficiency and effectiveness of the algorithm at a user given regularity threshold.