Data Science from Scratch, 2nd Edition

Chapter 17. Decision Trees

A tree is an incomprehensible mystery.

Jim Woodring

DataSciencester’s VP of Talent has interviewed a number of job candidates from the site, with varying degrees of success. He’s collected a dataset consisting of several (qualitative) attributes of each candidate, as well as whether that candidate interviewed well or poorly. Could you, he asks, use this data to build a model identifying which candidates will interview well, so that he doesn’t have to waste time conducting interviews?

This seems like a good fit for a decision tree, another predictive modeling tool in the data scientist’s kit.

What Is a Decision Tree?

A decision tree uses a tree structure to represent a number of possible decision paths and an outcome for each path.

If you have ever played the game Twenty Questions, then you are familiar with decision trees. For example:

• “I am thinking of an animal.”

• “Does it have more than five legs?”

• “No.”

• “Is it delicious?”

• “No.”

• “Does it appear on the back of the Australian five-cent coin?”

• “Yes.”

• “Is it an echidna?”

• “Yes, it is!”

This corresponds to the path:

“Not more than 5 legs” → “Not delicious” → “On the 5-cent coin” → “Echidna!”

in an idiosyncratic (and not very comprehensive) “guess the animal” decision tree (Figure 17-1).

Figure 17-1. A “guess the animal” decision tree

Decision trees have a lot to recommend them. They’re ...