Improved Positive and Negative Quantitative Association Rule Mining using SAM

Mining and discovering association from huge dataset is one of the common data mining technique, which help to extract interesting knowledge and find dependencies between items in the dataset. Several techniques and algorithms have been proposed to find dependencies between items with positive dependencies and those techniques don’t concentrate on negative dependency calculation. Certain algorithms initiated the findings of negative association rules, even though the techniques are effective, that is only considered the quality in rules. So there is a need of finding both positive and negative quantitative association rules from the dataset. This paper proposes a new technique named as Improved Positive and Negative Quantitative Association Rule Mining using SaM(Split and Merge). It is a new multi-objective based algorithm, which helps to mine a decreased set of positive and negative quantitative association rules rapidly. In addition, this proposal maximizes the following objectives such as improving precision, interestingness, performance and reducing the storage overhead. This also includes the split and merge algorithm for fast data management. in order to obtain set of rules which are interesting, easy to understand, suitable for decision making and provide good coverage of the dataset. The effectiveness of the proposed approach is validated over several real-world datasets.