josh.meanings.kmeans — org.clojars.joshua/josh.meanings 0.2.3

K-Means clustering generates a specific number of disjoint, non-hierarchical clusters. It is well suited to generating globular clusters. The K-Means method is numerical, unsupervised, non-deterministic and iterative. Every member of a cluster is closer to its cluster center than the center of any other cluster.

The choice of initial partition can greatly affect the final clusters that result, in terms of inter-cluster and intracluster distances and cohesion. As a result k means is best run multiple times in order to avoid the trap of a local minimum.

assign-clusters clj

(assign-clusters k-means-state)

calculate-objective clj

(calculate-objective k-means-state)

default-distance-fn clj

default-format clj

default-init clj

default-run-count clj

estimate-size clj

(estimate-size config)

Estimates the number of records in the dataset.

find-closest-centroid clj

(find-closest-centroid configuration)

initialize-centroids! clj

(initialize-centroids! k-means-state)

initialize-k-means-state clj

(initialize-k-means-state points-file k options)

Sets initial configuration options for the k means calculation.

k-means clj multimethod

k-means-seq clj

(k-means-seq dataset k & options)

Returns a lazy sequence of m ClusterResult.

recalculate-means clj

(recalculate-means k-means-state)

stabilized? clj

(stabilized? centroids-1 centroids-2)

K-means is said to be stabilized when performing an iterative refinement (often called a lloyd iteration), does not result in any shifting of points between clusters. A stabilized k-means calculation can be stopped, because further refinement won't produce any changes.

update-centroids clj

(update-centroids k-means-state)