Did not converge warning for k-means clustering in R

user_6123824 0
Viewed 594

I am trying to calculate optimum number of clusters in R. I am using the following code

library(factoextra)
library(tidyverse)
library(cluster)

wsCustomer <- read.csv(url("https://archive.ics.uci.edu/ml/machine-learning-databases/00292/Wholesale customers data.csv"))

#Converting Region and Channel columns ; replacing values by names
wsCustomer <- wsCustomer %>% mutate(Channel = ifelse(Channel == 1 , "HoReCa","Retail"),
                                    Region = case_when(Region == 1 ~ "Lisbon",
                                                       Region == 2 ~ "Oporto",
                                                       Region == 3 ~ "Others"))

head(wsCustomer)

df <- as_tibble(scale(wsCustomer[3:8])) 

# compute gap statistic
set.seed(123)
gap_stat <- clusGap(df, FUN = kmeans, nstart = 25,
                    K.max = 10, B = 50)

It is giving me the following warning

Warning message: did not converge in 10 iterations

How to get rid of this warning message?

cluster-analysis r
1 Answers

An answer from here

It means that the partition obtained is not stable (i.e. the algorithm did not converge toward an optimal solution). Indeed, a supplementary iteration will modify it significantly.

Default iter.max parameter is set to 10, which in your case it's not enough so just increase it by setting iter.max parameter:

gap_stat <- clusGap(df, FUN = kmeans, nstart = 25, K.max = 10, B = 50,iter.max=30)
user_61492110
Related