Constructing K-Means Models using Clusterers

We offer several different clusterers that implement LLoyd's algorithm to find optimal clusterings.

Name	Algorithm
MultiKMeansClusterer.COLUMN_TRACKING	high performance implementation that performs less work on later rounds
MultiKMeansClusterer.MINI_BATCH_10	a mini-batch clusterer that samples 10% of the data each round to update centroids
MultiKMeansClusterer.RESEED	a clusterer that re-seeds empty clusters

The COLUMN_TRACKING algorithm tracks the assignments of points to clusters and the distance of points to their assigned cluster. In later iterations of Lloyd's algorithm, this information can be used to reduce the number of distance calculations needed to accurately reassign points. This is a novel implementation.

The MINI_BATCH_10 algorithm implements the mini-batch algorithm. This implementation should be used when the number of points is much larger than the dimension of the data and the number of clusters desired.

The RESEED algorithm fills empty clusters with newly seeded cluster centers in an effort to reach the target number of desired clusters. You may use the apply method of the companion object MultiKMeansClusterer to create a cluster.

package com.massivedatascience.clusterer

object MultiKMeansClusterer {
  def apply(clustererName: String): MultiKMeansClusterer = ???
}