Practical Applications of Data Mining

7.4 Clustering ConCepts 285

make, marital status, and zip code are all valid choices for variables because

they all directly or indirectly affect the number of claims. On the other hand,

the inclusion of a variable such as the height or weight of an automobile

may adversely affect the outcome of the categorization because they are not

relevant to the problem. Therefore, a potential investigator has many choices

when deciding the strategy of the initial process. How many variables should

an investigator choose to adequately categorize an object? The answer is

the fewer the better to adequately address the problem since the inclusion of

irrelevant variables drastically affects the outcome. If a single variable will

bring about a straightforw ...