Clustering Flashcards
1
Q
What is inertia?
A
the sum of squared distances from each point to it’s assigned centroid
2
Q
How to choose the ideal number of clusters?
A
the elbow method
3
Q
Could we simply choose k by minimizing the inertia?
A
No, because inertia will always decrease as we increase the number of clusters.
4
Q
How to use the elbow method?
A
we’ll visualize the inertia against our choice in number of clusters.
In this visualization, we’ll look for the “elbow” of the curve, or the point where adding additional clusters doesn’t give us as big of a reduction in inertia.
