An Iterative Improvement Procedure for Hierarchical Clustering

We describe a procedure which finds a hierarchical clustering by hillclimbing. The cost function we use is a hierarchical extension of the  -means cost; our local moves are tree restructurings and node reorderings. We show these can be accomplished efficiently, by exploiting special properties of squared Euclidean distances and by using techniques from scheduling and VLSI algorithms.

• Type: Conference
• Year: 2003
• Where: NIPS
• Authors: David Kauchak, Sanjoy Dasgupta
• Keywords: Hierarchical Clustering, Hierarchical Extension, NIPS 2003, NIPS 2007, Squared Euclidean Distances