Data Mining Techniques for Segmentation - Data Mining Techniques in CRM: Inside Customer Segmentation - page 88

Figure 3.6 IBM SPSS Modeler recommended K-means Model options.
K-means is that it does not automatically provide an optimal clustering solution,
thus it typically requires many runs before coming up with the most efficient
grouping.
CLUSTERING WITH THE TWOSTEP ALGORITHM
The TwoStep algorithm involves a scalable clustering technique that can effi-
ciently handle large datasets. It incorporates a special probabilistic distance
measure that can efficiently accommodate both continuous and categorical input
attributes.