Exercise 1 Pruning The Boundary Decision Of The Nearest Neighbor Classifier Prove That When An Element Of The Training 1 (27.83 KiB) Viewed 50 times
This exercise is on page 227 of the followingbook: I don't understand the exercise very well, I'd begrateful if you could help me in detail with the proof.
Exercise 1 (Pruning the boundary decision of the nearest neighbor classifier) Prove that when an element of the training set has its neighboring Voronoi cells (called natural neighbors) of the same class, then this sample can be safely removed without changing the boundary decision of the NN classification rule. How to prune the training set for the k-NN rule?
Join a community of subject matter experts. Register for FREE to view solutions, replies, and use search function. Request answer by replying!