Page 63 - Data Science Algorithms in a Week

3

Decision Trees

A decision tree is the arrangement of the data in a tree structure where, at each node, data is
separated to different branches according to the value of the attribute at the node.
To construct a decision tree, we will use a standard ID3 learning algorithm that chooses an
attribute that classifies the data samples in the best possible way to maximize the
information gain - a measure based on information entropy.
In this chapter, you will learn:

What a decision tree is and how to represent data in a decision tree in example
Swim preference
In the section Information theory concepts of information entropy and
information gain theoretically first, then practically applying on example Swim
preference
ID3 algorithm constructing a decision tree from the training data and its
implementation in Python
How to classify new data items using the constructed decision tree in example
Swim preference
How to provide an alternative analysis using decision trees to the problem
Playing chess from the previous chapter and how the results of two different
algorithms applied may differ
Verifying your understanding at the exercise section when to use and when not
to use decision trees as a method of analysis
How to deal with data inconsistencies during decision tree construction in
example Going shopping