APPLICATION OF SUPERVISED LEARNING
UNSUPERVISED LEARNING
|
Question
[CLICK ON ANY CHOICE TO KNOW THE RIGHT ANSWER]
|
|
|
yes
|
|
|
no
|
|
|
Either A or B
|
|
|
None of the above
|
Detailed explanation-1: -Q6. For two runs of K-Mean clustering, is it expected to get the same clustering results? K-Means clustering algorithm instead converses on local minima, which might also correspond to the global minima in some cases but not always.
Detailed explanation-2: -While K-means shall always converge to at least a local minima (in sufficient number of iterations), its convergence to a global minima is not guaranteed. Consider the following example: In the below figure we have a set of blue points on the left side, and a set of red points on the right side.
Detailed explanation-3: -It sounds like an initialization issue. K-means generally needs some initial cluster assignment or set of cluster centers to start with. The two differing results might hence likely be two local minima of the function (minimal distances to class means) that k-means optimizes.