Perform a k-means clustering (with 3 clusters) of the one-dimensional set of points $1,1,2,3,4,7,8,8,12,14,16,21,22,24$ and with the inital point $=2$

To start I would select $3$ centroids that are as far away from one another as possible, e.g. $c_{1}=1, c_{2}=8, c_{3}=24$ and then calculate the distance from each point to each centroid. I would then assign each point to the cluster for which it is nearest to the centroid.

However I don't know what the question means by inital point $=2$?

  • 2
    $\begingroup$ I don't either know what they mean initial point 2. Initial point could be centroids like you say, which would be 3 values or initial sets which could be indexes or vectors of which of the values belong to each of the three for start. $\endgroup$ – mathreadler Apr 26 '17 at 10:18
  • $\begingroup$ You need a special rule in the K-means algorithm when one cluster is empty or when two clusters have the same centroid $\endgroup$ – reuns Apr 26 '17 at 22:00

It means you should run the K-Means algorithm and set all 3 initial centroids centers to be 2.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.