R cluster

Solution in R

As you can imagine, the solution in R us much shorter and requires much less computation on the user side. We first need to enter the data as a matrix or dataframe:

X <- matrix(c(7, 3, 4, 5, 2, 4, 0, 1, 9, 7, 6, 8), nrow = 6, byrow = TRUE ) X # display the coordinates of the points ## [,1] [,2] ## [1,] 7 3 ## [2,] 4 5 ## [3,] 2 4 ## [4,] 0 1 ## [5,] 9 7 ## [6,] 6 8

We now perform the k-means via the function with the point 5 and 6 as initial centers:

# take rows 5 and 6 of the X matrix as initial centers res.k <- kmeans(X, centers = X[c(5, 6), ], algorithm = "Lloyd" )

Unlike in the previous application with the dataset where the initial centers are randomly chosen by R, in this second application we want to specify which points are going to be the two initial centers.

For this, we need to set to indicate that that there are 2 centers, and that they are going to be the points 5 and 6 (see a reminder on how to subset a dataframe if needed).

The reason for adding the argument can be found in the usage of the R function .

how to group data in react js

how to cluster categorical data in r