Question

Which of the following is true for the nearest neighbor classifier (Select all that apply):
1 point
Partitions observations into k clusters where each observation belongs to the cluster with the nearest mean
Memorizes the entire training set
A higher value of k leads to a more complex decision boundary

Answer 1

Answer:

In KNN, finding the value of k is not easy. A small value of k means that noise will have a higher influence on the result and a large value make it computationally expensive. Data scientists usually choose as an odd number if the number of classes is 2 and another simple approach to select k is set k=sqrt(n).....

Answer 2

Answer:

Only correct answer is "Memorizes the entire training set"

Explanation:

The algorythm looks at the closest k-number of previous labeled observations and simply selects the most common one.

Therefore when we have large k value, we consider a lot of other observations which leads to less complex decision boundary. For example when we select k equal to the number of all previous observations (largest k we can choose), we would always just pick the most common class as our answer

Partitioning observations into clusters is true for k-means clustering = different algorythm