ABSTRACT

An association rule technique generally used to generate frequent itemsets from databases and generates association rules by considering each item in the datasets. However, the values of items are different in many aspects in a number of real applications, such as retail marketing, network log, etc. The difference between items makes a strong impact on the decision making in these applications. Therefore, traditional Association Rule Mining(ARM) cannot meet the demands arising from these applications. In this paper a new approach is introduced for computing profit weight of an item and generating frequent itemsets by minimum support threshold. The profit or the importance of the items in the itemsets is computed, based on the item subjective measures of characteristic through the proposed Global Profit Weight (GPW) algorithm using multi criteria decision making technique to improve the quality of output .