R/kmeans_cluster.R

In popiela/homework: Homework assignment

#' Clustering the respondents based on philosophy module using kmeans clustering method.
#' @param df
#'
#' @return
#' @export
#'
#' @examples
#' homework::kmeans_cluster(survey_data)
kmeans_cluster <- function(df)
{

  age <- df %>% dplyr::select(response_id,s_age)

  df <- df%>%
    dplyr::select(response_id , contains('m1_philosophy'))

  dm <- data.matrix(df[2:10])

  km <- kmeans(dm, 4 , nstart = 50)

  joined <- cbind(df , cluster = km$cluster)  %>%
    dplyr::left_join(age) %>%
    dplyr::mutate(cluster = paste0("Cluster ", cluster))%>%
    dplyr::select( response_id , cluster,s_age)

  Segment <- joined %>%
    dplyr::group_by(cluster , s_age) %>%
    dplyr::summarise(count = dplyr::n(), .groups = 'drop' )   %>%
    dplyr::ungroup() %>%
    tidyr::pivot_wider(names_from =  cluster ,  values_from = count)

  return(Segment)
}