Unsupervised Learning – K-Means

(you may now go back to the lecture hall)

Consider a K-means algorithm performed on the following data-points:

A1 = (2, 10) 
A2 = (2, 5) 
A3 = (8, 4) 
A4 = (5, 8) 
A5 = (7, 5) 
A6 = (6, 4) 
A7 = (1, 2) 
A8 = (4, 9)

Consider K=3 and the intial cluster centres (or centroids) being A1, A4, and A7. Use Euclidean distance.

What are the three cluster centroids after the first round of execution (iteration)? And the second? When is it time to stop new iterations? When you reach that point, your final answer (and password) will be the x coordinate of the (final) cluster center that started in A1, plus the y coordinate of the (final) cluster center that started in A4, plus the y coordinate of the (final) cluster center that started in A7.

Then go to https://kohan.uk/unsupervised-2/