JSPM

  • ESM via JSPM
  • ES Module Entrypoint
  • Export Map
  • Keywords
  • License
  • Repository URL
  • TypeScript Types
  • README
  • Created
  • Published
  • Downloads 25
  • Score
    100M100P100Q69527F
  • License MIT

A TypeScript implementation of AGNES, an agglomerative hierarchical clustering algorithm

Package Exports

  • dbvis-hc

This package does not declare an exports field, so the exports above have been automatically detected and optimized by JSPM instead. If any package subpath is missing, it is recommended to post an issue to the original package (dbvis-hc) to support the "exports" field. If that is not possible, create a JSPM override to customize the exports field for this package.

Readme

dbvis-hc

npm version Build Status codecov dependencies Status devDependencies Status License: MIT

A TypeScript implementation of AGNES, an agglomerative hierarchical clustering algorithm.

Install

Install with npm:

npm install --save dbvis-hc

This package requires module resolution by Node in tsconfig.json:

{
    "compilerOptions": {
        "moduleResolution": "node"
    }   
}

Usage

This implementation can be used to cluster generic objects, given a useful definition of distance and in case of the Centroid-Linkage, additionally a definition for aggregation.

Example:

    import { CentroidLinkage, Dendrogram, HierarchicalClustering } from 'dbvis-hc';

    const data = [ 10, 0.9, 1.0, 11, 1.1 ];
    // Distance function required by all linkage strategies 
    const distFunc = (a: number, b: number): number => Math.abs(a - b);
    // Aggregation function only needed by the centroid linkage strategy
    const aggrFunc = (v: number[]): number => v.reduce((acc, curr) => acc + curr, 0) / v.length;
    const hc = new HierarchicalClustering<number>(data, new CentroidLinkage(distFunc, aggrFunc));
    const rootCluster = hc.cluster();
    // Split cluster according to the dendrogram and a cut-off value
    const dendrogram = new Dendrogram<number>(rootCluster, 9.3);
    const clustersAsRefs = dendrogram.extractClustersAsRefs();
    // clustersAsRefs: [ [1.1, 0.9, 1.0], [10, 11] ]
    const clustersAsIds = dendrogram.extractClustersAsIds();
    // clustersAsIds: [ [ 4, 1, 2 ], [ 3, 0 ] ]

Linkage Strategies

This package provides common linkage strategies:

  • Single-Linkage
  • Complete-Linkage
  • Average-Linkage
  • Average-Group-Linkage
  • Centroid-Linkage

You can provide your own linkage method by subtyping the AbstractLinkage class.