27: Decision Trees (1 of 2)

So far our classification picture has been very general. We haven’t said anything about how our classifier might actually work; we’ve just said that given values for each of the features, it will render a prediction about what the label will be.

In chapters 27 and 28, we’ll study one particular algorithm for classification in machine learning: the decision tree algorithm. Not only does it make a good introductory technique because of its intuitive appeal, and not only can it classify pretty well in its own right, but it also serves as the basis for a more sophisticated, stateof-the-art classification method called “random forest” which we’ll explore in Volume Two of this series.

---