Calculate Within-Cluster Sum of Squares for Different k

Used for elbow method to determine optimal k

Usage

calc_wss(data, max_k = 10, nstart = 25)

Arguments

data

A data frame or tibble

max_k

Maximum number of clusters to test (default: 10)

nstart

Number of random starts for each k (default: 25)

Value

A tibble with columns k (number of clusters) and tot_withinss (total within-cluster sum of squares).

Examples

# \donttest{
wss <- calc_wss(iris[, 1:4], max_k = 6)
plot(wss$k, wss$tot_withinss, type = "b")

# }