Running head: CLUSTER ANALYSIS 1 Cluster Analysis Using SPSS Modeler Name Institution Date CLUSTER ANALYSIS 2 Cluster Analysis-K-means Using SPSS Modeler Introduction Cluster analysis using k-means was performed in SPSS modeler using Wholesale Customers dataset to determine what patterns emerge and how these patterns can be used to create specific client marketing profiles. The results the wholesale firm to understand the purchasing profiles of its clients. Therefore, the cluster used client type as well as region to classify customers based on product variables they purchase which were measured continuously. Methodology Five variables measured continuously were used as inputs to create three (3) clusters based on client type by generating distance fields. The variables include; Fresh: Annual spending (in $1000s) on fresh products, Milk: Annual spending (in $1000s) on milk products, Grocery: Annual spending (in $1000s) on grocery products, Frozen: Annual spending (in $1000s) on frozen products, Detergents Paper: Annual spending (in $1000s) on detergents and paper products. A distribution graph and a scatterplot were used whereby $KM-K-Means was placed to the X field variable, and $KMD-K-Means was placed to the Y field. For the frequency distribution graph, $KM-K-Means was placed as the primary field. The process stream in Figure 1 below. CLUSTER ANALYSIS 3 Figure 1: Process Stream for K-Means Cluster in SPSS Modeler Presentation of Findings Question 1 The cluster an ...
Purchase document to see full attachment