next up previous
Next: 3.5 ECH (Delaunay variant Up: 3 Algorithms Implemented Previous: 3.3 -MEANS

3.4 TB (\( k\)-MEDOIDS)

The TB clustering algorithm implements the TB heuristic to the \( k\)-MEDOIDS (or \( p \)-median) problem, as detailed in Section 3 of [1], originally proposed in [2]. It is similar to \( k\)-MEANS, except that it has the restriction that the cluster representatives must be points of the dataset.

It runs in \( \Omega (n^{2}) \) time, but gives substantially better quality clusterings than \( k\)-MEANS. It requires the desired number of clusters, \( k\), to be specified on the command line.

Kevin Pulo