CS6210: Document clustering in DLs
8
Distance Functions
wSingle Link -- O(n2)
*Distance = minimum document distance between 2 clusters.
wComplete Link -- O(n3)
*Distance = maximum distance between 2 clusters.
wGroup Average – O(n2)
*Distance = average document distance between 2 clusters.
wDistance function -- cosine measure
*Cosine(d1,d2) = (d1 • d2) / ||d1|| ||d2||
–
w