Class KMeans

java.lang.Object

org.episteme.core.mathematics.ml.KMeans

public class KMeans
extends Object

k-means clustering algorithm.

Partitions n observations into k clusters based on nearest centroid. Unsupervised learning for data grouping, pattern recognition.

Since:

1.0

Author:

Silvere Martin-Michiellot, Gemini AI (Google DeepMind)

Constructor Summary

Constructors

Constructor	Description
KMeans(int k, int maxIterations)

Method Summary

Modifier and Type	Method	Description
int[]	fit(Real[][] data)	Fits k-means model to data.
Real[][]	getCentroids()
Real	inertia(Real[][] data, int[] assignments)	Computes inertia (sum of squared distances to centroids).
int[]	predict(Real[][] newData)	Predicts cluster for new data points.

Methods inherited from class Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Constructor Details

KMeans

public KMeans(int k,
 int maxIterations)

Method Details

fit

public int[] fit(Real[][] data)

Fits k-means model to data.

Parameters:

data - n samples × d features

Returns:

cluster assignments for each sample

inertia

public Real inertia(Real[][] data,
 int[] assignments)

Computes inertia (sum of squared distances to centroids).

Lower is better. Used for elbow method to find optimal k.

getCentroids

public Real[][] getCentroids()

predict

public int[] predict(Real[][] newData)

Predicts cluster for new data points.