Mapping customer behaviors for segmentation

How can high dimensional data be visualized, so as view similarities, detect groups, and find outliers? In this example, we analyze customer data from a bank, and project records onto a 2D self-organizing map to achieve customer segmentation. We also explain how the same technique can be applied to other types of data.

Analyzed data:

The following anonymized variables were made available by a bank:

• Age
• Income
• Number of cars
• Number of children
• Sex
• Region (ex: city)
• Married
• Credit limit
• Opened a savings account
• Checking account balance
• Customer since (in months)
• Number of support calls
• Mortgage loan
• Defaulted on payment