@visactor/vmind/cjs/utils/cluster.js

<div align="center"> <a href="https://github.com/VisActor#gh-light-mode-only" target="_blank"> <img alt="VisActor Logo" width="200" src="https://github.com/VisActor/.github/blob/main/profile/logo_500_200_light.svg"/> </a> <a href="https://githu

"use strict";

function jaccardSimilarity(columnA, columnB) {
    if (columnA.length !== columnB.length) throw new Error("Columns must be of the same length");
    let intersection = 0, union = 0;
    for (let i = 0; i < columnA.length; i++) 1 !== columnA[i] && 1 !== columnB[i] || (union++, 
    1 === columnA[i] && 1 === columnB[i] && intersection++);
    return intersection / union;
}

function calculateClusterDistance(clusterA, clusterB) {
    let sumDistance = 0;
    for (const itemA of clusterA.children) for (const itemB of clusterB.children) {
        sumDistance += 1 - jaccardSimilarity(itemA.value, itemB.value);
    }
    return sumDistance / (clusterA.children.length * clusterB.children.length);
}

function calculateDistanceMatrix(data) {
    const n = data.length, distanceMatrix = Array.from({
        length: n
    }, (() => Array(n).fill(0)));
    let minDistance = 1;
    const distancePair = [ 0, 0 ];
    for (let i = 0; i < n; i++) for (let j = i + 1; j < n; j++) {
        const distance = calculateClusterDistance(data[i], data[j]);
        distanceMatrix[i][j] = distance, distanceMatrix[j][i] = distance, distance < minDistance && (minDistance = distance, 
        distancePair[0] = i, distancePair[1] = j);
    }
    return {
        distanceMatrix: distanceMatrix,
        minDistance: minDistance,
        distancePair: distancePair
    };
}

Object.defineProperty(exports, "__esModule", {
    value: !0
}), exports.agglomerativeHierarchicalClustering = exports.calculateDistanceMatrix = exports.jaccardSimilarity = void 0, 
exports.jaccardSimilarity = jaccardSimilarity, exports.calculateDistanceMatrix = calculateDistanceMatrix;

const agglomerativeHierarchicalClustering = (data, threshold = .4) => {
    const clusters = data.map(((v, i) => ({
        id: i,
        children: [ v ]
    }))), clusterMap = new Map;
    let hasMerged = !0;
    for (;hasMerged && clusters.length > 1; ) {
        hasMerged = !1;
        const {minDistance: minDistance, distancePair: distancePair} = calculateDistanceMatrix(clusters);
        if (minDistance <= threshold) {
            const mergedCluster = {
                id: clusters.length,
                children: [ ...clusters[distancePair[0]].children, ...clusters[distancePair[1]].children ]
            };
            clusters.splice(distancePair[0], 1), clusters.splice(distancePair[1] - 1, 1), clusters.push(mergedCluster), 
            hasMerged = !0;
        }
    }
    return clusters.forEach(((v, i) => {
        v.children.forEach((item => {
            clusterMap.set(item.id, i);
        }));
    })), {
        clusters: clusters,
        clusterMap: clusterMap
    };
};

exports.agglomerativeHierarchicalClustering = agglomerativeHierarchicalClustering;
//# sourceMappingURL=cluster.js.map