K-Means Clustering (Clusters.KMeans)

K-Means clustering partitions datasets into a chosen number of groups by minimizing within-cluster variance. Use the Clusters.KMeans implementation in @kanaries/ml to build fast, client-side segmentation pipelines for browser or Node.js applications.

constructor (n_clusters: number = 2, opt_ratio: number = 0.05, initCenters?: number[][], max_iter: number = 30)

const X = [
    [0, 0],
    [0.5, 0],
    [0.5, 1],
    [1, 1],
];
const sampleWeights = [3, 1, 1, 3];
const initCenters = [[0, 0], [1, 1]];

const kmeans = new KMeans(2, 0.05, initCenters);

const result = kmeans.fitPredict(X, sampleWeights);

Practical guide: K-Means Clustering (KMeans) in JavaScript and TypeScript

KMeans partitions data into k centroid-based groups for segmentation and coarse structure discovery.

When to use K-Means Clustering (KMeans)

• You can estimate a reasonable cluster count from domain knowledge.
• Clusters are roughly spherical in normalized feature space.
• You need fast, repeatable clustering for product analytics workflows.

Implementation workflow

1. Normalize numeric features and choose an initial value of k.
2. Train with multiple seeds when possible to reduce local-minimum sensitivity.
3. Validate cluster usefulness with silhouette score and business outcomes.

JavaScript deployment notes

• Prefer feature scaling for distance-based and gradient-based algorithms to improve stability.
• In browser apps, run heavy training in Web Workers to keep UI interactions smooth.
• Keep a simple baseline from the same module as a fallback model for comparison.

Search intents this page targets

• K-Means Clustering (KMeans) JavaScript
• K-Means Clustering (KMeans) TypeScript
• K-Means Clustering (KMeans) browser machine learning
• @kanaries/ml K-Means Clustering (KMeans)