Decision trees

A simple predictive model maps the outcomes of an item to the input data. It is a popular predictive modeling technique, which is used commonly in the industry:

Decision tree models are basically of two types:

  • Classification trees: These refer to dependent variables that take a finite value. In these tree structures, branches represent the rules of the features that lead to the class labels, and leaves represent the class labels of the outcome.
  • Regression trees: When dependent variables takes continuous values, then they're called regression trees.

Let's take an example. The following data represents whether you should play tennis or not, based on the overall outlook of weather, humidity, and wind intensity:




Get Mastering Python for Data Science now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.