Performance prediction using modified clustering techniques with fuzzy association rule mining approach for retail

Abstract

Clustering a group of data based on the related components and its similarity, using fuzzy association rule mining is a key to implement data mining in Soft Computing. Traditional way of clustering is only one object is assigned to a cluster, when it is overlapped and had more cluster to an existing object then fuzzy logic is used. Here Modification in the criteria value for membership value of clustering point of an object. Proximity measure is applied in Box metric equation. An analysis is made on retail database this makes better enhancing way to predict the sale and performance based on association rule mining using fuzzy model. The category has been selected in different number of products grouping the products based on the need from customer and Integration of Apriori Model to multi membership and multiple support approach for sale performance prediction.