Chapter 7. Deriving New Fields
We have now spent several chapters on the data preparation phase of data mining. We have learned how to select and sort cases, how to identify and remove duplicate cases, how to reclassify and filter fields, and how to combine different types of data files. In this chapter you will learn how to create additional fields by:
- Deriving fields as formulas
- Deriving fields as flags
- Deriving fields as nominals
- Deriving fields as conditionals
A very important aspect of every data mining project is to extract as much information as possible. Every project will begin with some data, but it is the responsibility of the data miner to gather additional information from what is already known. In our experience, this can be the most creative and challenging aspect of a data mining project. For example, you might have survey data, but this data might need to be summed for more information, such as a total score on the survey, or the average score on the survey, and so on. In other...
