Group data using the K-means clustering algorithm in JavaScript

1. As no initial centroids are provided, start by using the first k data points as the initial centroids, using Array.prototype.slice().
2. Initialize the distances array to store the distances between each data point and each centroid, as well as the classes array to store the cluster assignments for each data point.
3. Use a while loop to repeat the assignment and update steps as long as there are changes in the previous iteration, as indicated by the itr variable.

const kMeans = (data, k = 1) => {
  const centroids = data.slice(0, k);
  const distances = Array.from({ length: data.length }, () =>
    Array.from({ length: k }, () => 0)
  );
  const classes = Array.from({ length: data.length }, () => -1);
  let itr = true;

  while (itr) {
    itr = false;

    for (let d in data) {
      for (let c = 0; c < k; c++) {
        distances[d][c] = Math.hypot(
          ...Object.keys(data[0]).map(key => data[d][key] - centroids[c][key])
        );
      }
      const m = distances[d].indexOf(Math.min(...distances[d]));
      if (classes[d] !== m) itr = true;
      classes[d] = m;
    }

    for (let c = 0; c < k; c++) {
      centroids[c] = Array.from({ length: data[0].length }, () => 0);
      const size = data.reduce((acc, _, d) => {
        if (classes[d] === c) {
          acc++;
          for (let i in data[0]) centroids[c][i] += data[d][i];
        }
        return acc;
      }, 0);
      for (let i in data[0]) {
        centroids[c][i] = Number.parseFloat(
          Number(centroids[c][i] / size).toFixed(2)
        );
      }
    }
  }

  return classes;
};

kMeans([[0, 0], [0, 1], [1, 3], [2, 0]], 2); // [0, 1, 1, 0]

1. As no initial centroids are provided, start by using the first k data points as the initial centroids, using Array.prototype.slice().
2. Initialize the distances array to store the distances between each data point and each centroid, as well as the classes array to store the cluster assignments for each data point.
3. Use a while loop to repeat the assignment and update steps as long as there are changes in the previous iteration, as indicated by the itr variable.
4. Calculate the Euclidean distance between each data point and centroid using Math.hypot(), Object.keys(), and Array.prototype.map().
5. Use Array.prototype.indexOf() and Math.min() to find the closest centroid for each data point.
6. Update the cluster assignments in the classes array and check if any changes were made.