1
Student’s Name
Course Title
Professor
Date
, 2
EXERCISE 1: [20 POINTS]
1. Compute the minimum number of clusters for an expert who needs to explain at
least 80% of the variation capture and explained by the kMeans clustering
machinery?
I used elbow method to determine the minimum number of clusters. The idea behind the
elbow method is that as you increase the number of clusters, the TWCSS will initially
decrease rapidly as the clusters become more specific and better match the underlying
structure of the data. However, at some point, adding additional clusters will no longer
significantly reduce the TWCSS. This point is known as the "elbow" and the number of
clusters corresponding to this point. In our case the minimum number of cluster is k = 5.
Because at this point adding additional cluster did not significant reduce TWCSS. At this
point we have the lowest reduction which is 1455.9. Then from there it flattens out
Student’s Name
Course Title
Professor
Date
, 2
EXERCISE 1: [20 POINTS]
1. Compute the minimum number of clusters for an expert who needs to explain at
least 80% of the variation capture and explained by the kMeans clustering
machinery?
I used elbow method to determine the minimum number of clusters. The idea behind the
elbow method is that as you increase the number of clusters, the TWCSS will initially
decrease rapidly as the clusters become more specific and better match the underlying
structure of the data. However, at some point, adding additional clusters will no longer
significantly reduce the TWCSS. This point is known as the "elbow" and the number of
clusters corresponding to this point. In our case the minimum number of cluster is k = 5.
Because at this point adding additional cluster did not significant reduce TWCSS. At this
point we have the lowest reduction which is 1455.9. Then from there it flattens out
