20.

(a) Draw a kd-tree for the data of Figure 7.1. The topmost feature to split on should be the one that most divides the examples into two equal classes. (Do not split on UserAction.) Show which training examples are at which leaf nodes.

(b) Show the locations in this tree that contain the closest training examples to a new case where the author is unknown, the thread is new, the length is long, and it was read at work.

(c) Based on this example, discuss which examples should be returned from a lookup on a kd-tree. Why is this different from a lookup on a decision tree?