The method works as follows.

1. For a data set with dimensionality, d, compute the variance

of data in each dimension (column).

2. Find the column with maximum variance call it cvmax

and sort it in any order.

3. Divide the data points of cvmax into K subsets, where K is

the desired number of clusters.

4. Find the median of each subset.

5. Use the corresponding data points (vectors) for each

median to initialize the cluster centers.-The method works as follows.

1. For a data set with dimensionality, d, compute the variance

of data in each dimension (column).

2. Find the column with maximum variance　call it cvmax

and sort it in any order.

3. Divide the data points of cvmax into K subsets, where K is

the desired number of clusters.

4. Find the median of each subset.

5. Use the corresponding data points (vectors) for each

median to initialize the cluster centers.