Machine Learning with scikit-learn Quick Start Guide

Building the decision tree

Now that we have the labeled dataset, we can create a decision tree, in order to convert the unsupervised machine learning problem into a supervised machine learning one.

In order to do this, we start with all of the necessary package imports, as shown in the following code:

from sklearn.tree import DecisionTreeClassifierfrom sklearn.externals.six import StringIOfrom IPython.display import Imagefrom sklearn.tree import export_graphvizimport pydotplusfrom sklearn import tree

Next, we rename the target column to a name that is appropriate (when we merged the target labels created by the k-means algorithm, it produced 0 as the default name). We can do this by using the following code:

#Renaming the target df['fraud'] ...