Learning Data Mining with Python - Second Edition

Creating good features

Simplification due to modeling is a key reason we do not have data mining methods that can just simply be applied to any dataset. A good data mining practitioner will need, or obtain, domain knowledge in the area they are applying data mining. They will look at the problem, the available data, and come up with a model that represents what they are trying to achieve.

For instance, a person's height feature may describe one component of a person, such as their ability to play basketball, but may not describe their academic performance well. If we were attempting to predict a person's grade, we may not bother measuring each person's height.

This is where data mining becomes more art than science. Extracting good features ...