$\newcommand{\ones}{\mathbf 1}$

Clustering and k-means

The goal of clustering a set of vectors is to
  1. choose the best vectors from the set
    Incorrect.
  2. divide them into groups of vectors that are near each other
    Correct!
  3. determine the nearest neighbors of each of the vectors
    Incorrect.

Check below all that apply. The $k$-means algorithm

The choice of $k$, the number of clusters to partition a set of vectors into
  1. is a personal choice that shouldn't be discussed in public
    Incorrect.
  2. depends on why you are clustering the vectors
    Correct!
  3. should always be as large as your computer system can handle
    Incorrect.