Practical Applications of Data Mining

7.3 Clustering proCeDures 283

are the result of adding extra points, known as noise, to both Figure 7.1(a)

and Figure 7.1(b), respectively.

In general, clustering is situation-specific and often exploratory. For insta-

nce, people can be grouped by gender, age, address, education, occupation,

earned income per year, number of dependents, and so on. If a management

team for a real estate company wants to decide how many buildings at various

price levels should be constructed, potential customers may be better grouped

according to earned income per year rather than age, because decision making

for a home purchase is more likely to depend on annual income. As another

example, for an insurance company to set appropriate driver’s insurance pre ...