k mean clustering on ip address

I attached ip address data file. I want to apply k mean clustering on ip address data, how to do it. as ip address has 4 numbers can it possible to map data in 4 dimension and form cluster.

[Image Analyst]

It should be able to. Make the array into 4 columns with as many rows as you have ip addresses. How many clusters did you want, k? And do you want k clusters on the 4-D array as a whole, or k clusters on each set of numbers separately/individually?

Answer 1

[Walter Roberson]

Map each entry into a row vector of a 2D array and cluster on the 2D array that results.

[Jay Hanuman]

I don't understand, "each entry into a row vector of a 2D array" means if data has

10.0.11.4 
10.0.7.4 
10.0.8.4
10.0.14.4

then it should have make like this

100 114 
100 74  
100 84 
100 144

[Walter Roberson]

 [10 0 11 4;   %10.0.11.4 become one row of this array
  10 0  7 4;   %10.0.7.4 became another row of this array
  10 0  8 4;   %10.0.8.4 become another row of this array
  10 0 14 4]   %10.0.14.4 became the last row of this array

The result is a 2D array, not a scalar and not a vector and not something with 3 dimensions.

[Jay Hanuman]

but how to cluster on the 2D array

[Walter Roberson]

https://www.mathworks.com/help/stats/kmeans.html

"idx = kmeans(X,k) performs k-means clustering to partition the observations of the n-by-p data matrix X into k clusters, and returns an n-by-1 vector (idx) containing cluster indices of each observation. Rows of X correspond to points and columns correspond to variables."